RF_RES_2_SUDHA.pdf
Media
- extracted text
-
September 21, 1995 Revised LKB, revised Dec 4/7 GK
Theories as an independent chapter
Intervention Mapping: Designing Theory and Data Based Health Education
Programs
CHAPTER 2: USING THEORIES, EXISTING EMPIRICAL EVIDENCE AND NEW
DATA
INTRODUCTION
Objectives for Chapter 2. The reader will be able to: 1) Pose planning
questions in ways that facilitate finding answers from theory, the literature and
RF_RES_2_SUDHA.pdf
*•
paillde uaAijp-Ajosqi uiojj uiajajjip si aousios lejoiAeipq paiidde uaAijp-iuaiqojd
•Buiajos waiqojd Aue oi ApoaJip amquiuoo oi aiunsse iou pjnoqs mq luaiu
-dopAQp Ajoaqj oj smqiJiuoo him aoqoejd ui Ajoaqi b Buiisai xsquaios e zpaiejs
A||npojo^ qjo|/\] ■aoijoejd ui Ajoaqi e jsa; 01 siubm oqM auo ueqj >|sei luajajjip
b seq tuaiqojd e joj uopnjos e puij 01 siubm oqM ispuaios lejoiAeqoq v
3ON3IOS “1VHOIAVH39 dSFIddV N3AIHO-IAI33aOHd
new research 2) Describe the difference between problem-driven applied
iiZHOHV3s Advaan inoav ivHMiZi
behavioral science and theory-driven applied behavioral science. 3) Use the
■[tz66 L '’IB
process of problem definition, provisional explanations, theory, and new data to
address questions related to determinants of behavior, methods for intervention.
and implementation of programs 4) Use the issue, construct and general
theories approaches to accessing theory for the solution of health education
ua||n|/\nc-t766l 'saBpan
The health educator reader will need to learn the use of tools, such as
available empirical literature, appropriate theories, and additional research data
to answer planning questions. These questions include defining behavior,
finding determinants of behavior, differentiating the target population, selecting
methods, creating strategies, and developing implementation plans. There is a
JQdooo] [9861 'UMIO Al saBpanl sisAieue-eiaiu pue
mbiabj ajmejaiii pue [noo|o/y\][8861 'Ja6naj>|]8861 'wnequaaJ9][886 I
'ueB
-jo|/\|][886L 'ueijjjaifll] [966 L '8>|B1S][^86 L'UBUJJ9AnH V sa|i|/\i][6Z.6 L 'S8|!l/Ml
uewjaqn|-|][ 166 I 'liqioo
ssneJis]t766 L 'U!A][986 L 'UIA11066 I
'uonej] qojeasaj aAiiBiijenb '[aBeg
AjnqMON 'qoeojddv oiieiuaisAs
[fc66 l's8|!l/\l
problems.
saiJA 801 [£86 L'zsjiiubh
V :uo!ien|EA3 6861 '3HueiuaaJd Hd !ssoy][
JOspui/\/\][9861 'siMaq
uaajQjuo
-ijeniBAa pue qojeasaj aApeiquenb pue aAiiemenb joj saojnos jaqio qnsuoo
ot japeaj aqi aBejnooua bm pue ajaqMasja A|aAisuajxa paquosap uaaq aAeq saio
-uaiadiuoo leuoissajojd asaqj. uaqiaBoi sassaoojd aajqi aqi jo asn aqi ajoujojd
wealth of information available in the existing literature on any health education
8M asneoaq ejep leuopippe Buijoanoo uo pue ajniejaiii qojeasaj aqj Buimbiabj
topic. Advanced methods for gaining information from libraries give access to
pue BuiqojBas uodn A||9uq qonot o/y\ •stuejBojd uoqeonpa qi|Baq jo Buiuueid
systematic experience of other researchers and practitioners, and what we can
not find directly in the research literature, we may find in theory.
At times we will have to supplement the use of theory and the research
aqi ui sauoaq; aieudojdde Buisn pue Buipuij uo snooj om jaideqo siqi u|
•sdnojfiqns sii pue uoiiB|ndod tafijet jeinoijjed jno pueisjapun oj ejep
paau os|e a/y\ -uoiienjis jeoiioejd aqi ui sidaouoo leoiiajoaqi }o Buiueaiu (enioe
aqj Ajioads 0} pue 'sjaMsue jeuoisiAOJd |0 Aijiiqisneid aqj >|oaqo oj 'waiqojd aq)
i
—
auipp ox qojeasaj Mau paau Aeiu a/y\
•qojeasaj mqu luoj^ etep qjiM ajniBJOjii
V
c
driven applied behavioral science assumes that, given empirical support, almost
behavioral science. In Intervention Mapping we are always working from a
all theories are useful within the conditions that limit the application of the
problem-driven perspective. Choices have to be made in the process of develo-
theory (McGuire, 1985; 1991).
ping an intervention and theories are one tool to enable us to make better
choices.
A MODEL FOR APPLYING THEORY
Compared to the huge amount of literature about theories in social
"All theories are right"
sciences, useful literature about the systematic problem-driven application of
The application of theories can be very useful from a practical point of view.
theories is minimal. The English literature offers the book of Lave & March
However, a one-sided focus on one, or only a few theories may lead to sugge
(1975), which describes the application of sociological models in particular, and
stions that may not contribute to a reduction or solution of a practical problem,
theJDutch Hterature offers P. Veen (1985) 'Applying social psychology'.
or moreover, that may be counterproductive. A thorough analysis of the
Problem-driven applied behavioral science refers to scientific activities that
practical problem is a first and essential step in problem-driven applied behavio-
focus on changing a problem by using a transtheoretical behavioral science
9
ral science. Researchers as well as practitioners should not restrict themselves
approach. Although theories are used, the main focus is on problem solving and
to one theory but they should look at all the different aspects of the problem.
the criteria for success are formulated in terms of the problem. Possible contri
Before potential theoretical frameworks are selected, we have to answer
butions to theory development is a useful but unnecessary side effect.
questions such as: what is the problem, why is it a problem, who's problem is
In the
next section, we will particularly focus on Veen's approach by applying it in the
it, what are possible causes. A careful analysis of the practical problem may
field of health education and promotion.
prevent us from spending time on irrelevant theories, irrelevant problems, or
The systematic approach 'from problem to solution' comprises four steps:
problems that are not in the area of behavioral science (psychology, sociology)
D
Problem definition or clarification
and that need approaches from other disciplines (economy, law, engineering).
2)
Formulation of provisional explanations
3)
Application of theoretical explanations plus additional research to formulate
A second important aspect of problem-driven applied behavioral science is
the selection of appropriate theories. To find out whether a theory is relevant
final explanations
for a given practical problem, we have to specify the conditions 1) that allow
4)
Formulation of provisional solutions
the theory to make predictions, and 2) that are necessary for theoretical
5)
Application of theoretical explanations plus additional research to formulate
concepts to be relevant (see Ajzen, 1988, p. 138-142; McGuire, 1991). More
final solutions
than in theory- driven fundamental and applied behavioral science, problem
4
3
assessment and intervention[Bartholomew et al. 1988; Ban.
.omew et al,
1990; Bartholomew et al. 1991]. Therefore, we had to know exactly what we
6)
Implerr.
ation and evaluation.
Figure 2.1 represents the process of applying theory for problem solving
Insert Figure 2.1: The process of applying theories
meant by coping with chronic illness or coping with cystic fibrosis. We posed
the following questions. Problem'. Cystic fibrosis is a serious disease that
places many burdens on parents and their children.
The disease must require
From Problem to Problem Definition
In general health promotion planning models, we use the concept
tremendous resources for coping and adjustment. Questions to the explanation:
'problem' for health-related problems such as 'There is excess mortality and
How do parents and children cope with these burdens of chronic illness"? and
morbidity from car accidents involving children. What can be done about this
"How are some coping strategies more successful than others?". Questions to
problem?' However, In Intervention Mapping, the concept of problem compri
the solution: "How can we teach children and parents adequate coping skills?"
ses more than health problems and refers to specific health education problems
These questions are answerable with theory, existing evidence and new data.
involving definition of behavior related to health problems, determinants of risk
Over time the focus may shift further in the intervention mapping
and health promoting behavior, the nature of effective interventions to change
process, when we understand more of the problem, the explanations and the
behavior, and plans for implementation of programs.
solutions. In that case the attention shifts to more specific questions about
Insert Figure 2.2: Problem definition at different levels of IM
solutions (or interventions): "What methods can help us teach parents flexible
The most practical way to define a problem is to indicate the problem,
coping skills?"
followed by the question to the explanation and finally, the question to the
The questions may also be different for subpopulations: "Why can some
solution. For example, Problem: 'A lot of serious accidents happen with
children learn management skills while others cannot?" or "How can we ensure
children in cars even though this can be avoided by using child restraint devices
that each child learns as many management skills as possible". Stage of
(CRDs). Question to the explanation: Why is it that parents do not use CRDs?
development is an obvious answer to this question and, consequently, the
and Question to the solution: How can we ensure that they will?'. In this case
target group was differentiated into early childhood, school age, and adolescen-
the question to the problem is formulated in terms of determinants of behavior
ce.
and the question to the solution in terms of intervention methods. Here is
Finally, the problem will shift to the implementation process, Problem:
"Some health professionals do not have the motivation or the skills to help
parents learn coping skills". Question to the explanation: "What influences
6
another example.
In the Cystic Fibrosis Family Education Program (Chapter 12) we wanted
to include both medical and coping self-management behaviors in our needs
5
*
What is the problem?
health profei
Why is it a problem?
solution’. "How can we motivate and train health professionals to help parents
To whom is it a problem?
learn coping skills?" Again, both questions can be answered by using theories,
What are the aspects of the problem?
emprical evidence, and additional data.
nals to help parents learn coping skills?" Questions to the
What may be the causes of the problem?
Is the problem likely to be resolved?
Is it desirable to solve the problem?
Although the above questions are concerned with, among other things,
In practice, a common mistake in defining a problem is that the problem
is accepted the way it has been offered by the person or organization presenting
it. An example: In 1980, one of the authors was involved in a campaign about
explanations and solutions, they are not meant to offer adequate answers at
road safety in a small Dutch town. The reason for this campaign was that it had
this stage, but simply to clarify the problem. For example, if a solution of a
been noticed that, in this town, relatively more accidents took place at intersec-
problem appears to be quite obvious, it may give cause for examining why this
tions with main roads than in similar towns. The organizations responsible
solution had never been chosen in the past. It is possible that inside the
allowed the problem to remain defined in health terms, i.e. there is an excess of
organization responsible for solving the problem, resistance has occurred against
mortality and morbidity resulting from traffic accidents. The developers of the
the most obvious solution. This puts things in another light, and calls for
campaign regarded public education as the appropriate solution and started a
redefining the problem to include questions to the solution.
large local campaign about right-of-way rules. They failed to asked additional
The first step in the process of clarifying the problem is to go to the
questions about the accident problem, ie. questions of behavior and determi-
literature to find what others know about the problem. We look at our specific
nants. In the middle of the campaign, the health educators discovered that the
problem such as coping with cystic fibrosis, and we look at similar problems
majority of all right-of-way accidents happened only at one point: the exit of the
such as coping with other chronic diseases of childhood. We also might look
car park near the supermarket. The problem could have been solved simply by
for literature on processes similar to coping, processes such as adjustment.
pruning some bushes or introducing a traffic light. This example illustrates how
This first step of going to the literature will remain a first step through all the
important it is to make a careful and thorough analysis of the given problem.
stages of bringing theory and research to bear on a problem.
An Example - Problem to Problem Definition - CRD Use
The Dutch Foundation for Traffic and Safety has a long history of
educating parents of young children to promote the consistent use
of child restraint devises (CRD) in cars to protect young children
8
To facilitate the problem analysis, Veen suggests a number of questions
to clarify the problem. Although these questions overlap considerably, they
provide good insight into all aspects of the problem.
7
example, by taking the views of different people involved with it, by generali
zing to similar problems, and by narrowing the question to certain populations
and situations. Reversing perspectives may generate new ideas, an extreme
example of which is: How can the problem be increased, how would we try to
reduce the use of condoms?
Insert Table 2.1: Provisional Explanations for Lack of Condom Use
against the potentially harmful consequences of an accident.
However, there were still a high number of car-accidents in which
children were the victim (problem). The Foundation decided to find
out why their educational programs did not work (question to the
explanation) and how they could be improved (question to the
explanation). Thinking about that problem and realizing that no
needs assessment on determinants had ever been done, our own
definition shifted somewhat from program failure to determinants
analysis; What are the reasons parents do not always use CRD's
(including program failure; question to the explanation) and how
can we promote the use of CRD's (not necessarily through a
campaign; question to the solution).
among Adolescents
Before brainstorming we will have gone to the literature if we have not
done so when defining the problem.
In formulating these provisional explanati-
Formulating Provisional Explanations
The step of formulating possible explanations of a problem is a creative
ons, behavioral scientists always use specific theoretical knowledge, whether
process which primarily involves free association and brainstorming in response
consciously or not. It is unavoidable to do so in this stage, but it should not
to concepts in the problem definition [Veen, 1985]. It is important to start with
hinder an open approach.
as many explanations as possible in response to a question. At the end, poor
CRD Example - Formulating a Provisional List
What could be the reasons that parents do not use CRDs?
1.
They may, of course, not have been exposed to information
and never have heard of CRDs. It is also possible that they
do not know that CRDs are especially meant to avoid serious
injury from accidents.
2.
They may think that they can prevent serious injury in
other ways such as by tightly holding the child.
3.
They may underestimate their risks Perhaps they think they
do not run any risks themselves, for example, because they
do not expect themselves to get involved in an accident or
because they expect to be able to hold the child in case of
an accident (risk perception).
4.
There may be other things involved, for example if parents
find approved CRDs too expensive or that they take too
much space (practical objections).
5.
Obviously, it is also possible that parents have actually
bought a CRD, but do not use it consistently, for example,
because they have two cars. Or because various children
have to be transported at the same time (difference be
tween acquisition and use).
6.
Educational level.
10
explanations may be dropped, but it is definitely not good to get stuck on one
single explanation too soon.
Often brainstorming with several people from different disciplines
concerned with the problem or on a program development team will lead to new
points of view. In brainstorming we state the question and then generate as
many answers as possible without editing or criticising our work. Recently, a
work group in one of our health education methods classes generated a list of
provisional explanations for the determinants question "Why do adolescents fail
to use condoms during sexual intercourse?" (See Table 2.1). The list contains
predisposing, enabling and reinforcing factors, and so-called contextual factors.
Contextual factors are more distal determinants that seem to be important but
are relatively difficult to change. The students looked for determinants, for
9
The next step is to reinforce provisional explanation^ wy theoretical
We ndve just formulated a number of provisional explanations why
foundation and to acquire additional information through research. After all,
parents would not use CRD's and for a possible failing of the campaign. At this
health education theories are the outcome of many years of thinking about
moment, there is no reason to favor one explanation over another. However, in
behavior and behavioral change. What we will do here is apply this knowledge
the following steps, we begin to take into account two criteria for good
and experience to our question. This will yield new explanations, cause us to
explanations: an explanation should describe a process and it should be plausi-
delete some explanations from the provisional list, and raise new questions to
ble.
be answered by means of additional research.
Veen (1985) has suggested three approaches to searching for theories to
A process explanation provides an answer to the question why? A
process explanation provides insight into the relations among variables and
match with provisional explanations, the subject-related, construct-related, and
between the variables and the problem or behaviour. The explanation that
general theories approaches. Although a clear distinction between these three
'higher educated parents use CRDs more often than less educated parents' does
approaches does not exist, they may yield different explanations to a provisional
not reflect a process yet. A process explanation might for example be that
list.
higher educated parents are relatively more convinced of the use of CRDs or
With the subject-related approach we search for theories (as well as empiri-
that they are relatively less concerned about the high price of these seats.
cal data) through the subject or issue. For example the students working on
Continuously asking 'Why?' helps to formulate process explanations. In this
condom use question would search for theories in literature specifically related
respect, a useful aid is to represent the explanation in a schematic model
to condom use. We would approach the literature through the subject of
consisting of boxes with arrows between them (Earp & Ennet, 1991).
condom use and we would find studies using the Health Belief Model [Lux &
plausible explanation means that it has to be examined with common sense,
Petosa, 1994 (HEQ 21-4)]) or Ajzen's Theory of Planned Behavior [Basen-
and survive. If we know that in health centres, parents receive consistent health
Enquist et al, X; Schaalma et al, 1993]. If that action is clearly inadequate, we
education about child restraint devices, most parents being unaware of the
could also look for theories in literature on contraception, as a related behavior
existence of CRDs would not be a very plausible explanation. However, it would
and/or for literature on risk-taking behavior by adolescents as a more general
be if a special group of parents were involved, for example imigrant parents
subject.
having difficulties with the language and consequently with health care facili-
The construct-related approach works from the constructs on the
provisional list. As we mentioned before, some of the items on the list are
12
ties.
Enhancing the Provisional List With Theoretical Explanations
11
A
safety, risks, risk protection etc. We look not only for err.r <cal
data but also for theories and we find that seat belt use has been
explained by , such as the Protection-Motivation Model [R. W.
Rogers, 1983; cite specific articles seat belt????]. Health Belief
Model [Janz & Becker, 1984; site specific articles???] or risk
perception theories [Weinstein, 1989; ???) and by regulation!???].
probably simiL .o, or remind us of certain theoretical constructs. For example,
confidence on the student's provisional list for condom use (table 2.1) is similar
to the Social Cognitive Theory construct self-efficacy. Knowledge of HIV and
STD risks may be related to perceived susceptibility and perceived seriousness
Construct-related approach
The list of provisional explanations mentioned a possible underesti
mation of personal risk. That construct leads us to risk perception
models, such as Weinstein's (X) Precaution Adoption Theory,
which explain why some people wrongfully think they do not run
any risk. People tend to underestimate risks when they think they
control the situation. A parent driving the car may believe he or she
controls the child's risk, as does the parent who will hold the child
in case of an accident. In addition, people always belief they run
less risks than others (unrealistic optimism), partly because they
have a stereotyped perception of parents who actually run risks
and partly because they overestimate their efforts to take caution
compared to what other parents undertakefVan der Pligt et al.,
1993]. Informal interviews with parents supported these theoretical
predictions. Explanations of the process of unrealistic optimism
suggest learning objectives such as that parents should be taught
that other parents also do take measures to protect their children,
and that car-accidents are partly uncontrollable and unpredictable.
As a result, parents are supposed to become more aware of the
real risks and consequently will use the CRD more consistently
(performance objectives).
from the Health Belief Model or outcome expectancies from Social Cognitive
Theory again. Working with the construct-related approach also means that we
apply the theory fully, meaning that most of the time a theory will have more
constructs than the one construct that leads us to it. For instance, when a
construct of outcome expectancies leads us to Social Cognitive Thory, we will
also apply other constructs from the theory, such as self-efficacy expectations
and observational learning (modeling).
In the general theories approach we look at our question through the
lense of a determinants theory or change theory and we think about the
usefulness of the specific constructs in that theory to our question. If our
question concerns determinants of behavior as in the case of condom use we
Other risk perception models such as the Health Belief Model [X] and
Protection Motivation Theory [X] cover more than just risk perception,
and indicate under what circumstances risk perception leads to adequate
action. They add the constucts of behavioral effectiveness and selfefficacy expectations. The threat of risk motivates parents to acquire and
use a CRD provided that they are convinced that such a seat constitutes
an effective means of protection (behavioural effectiveness) and that they
believe they are able to acquire such a seat and use it consistently (selfefficacy).
may go to the Theory of Planned Behavior, for example, and consider subjective
norms, attitudes, self-efficacy expectations and behavioral intentions [Ajzen,
1986]. Clearly, the construct and general theories approaches are limited by
the number of theories with which a planner is familiar, and we devote a next
chapter to a brief review of theories commonly used in health education.
Other constructs in the list of provisonal explanations have to do
with the costs and inconvenience of CRD's. Consequently, we
think of the cost-benefit balance as it occurs in the description of
attitude [Ajzen, 1988]. This attitude theory in turn is part of a
model of behavioural determinants which suggests three types of
determinants: attitude (risk perception, behavioural effectiveness,
costs and inconvenience included), social influence and self-effica
cy. In our example of CRDs, it can be noticed that the provisional
explanations consider neither social influence nor self-efficacy.
Subject-related approach
Leaving our provisional list for a while, we use the issue or subject
approach and review the literature concerning the use of CRDs,
related behaviors such as seat belts use, and concepts such as
14
13
Example - Applying Theory to the Problem of CRD Use
Additional research
We may state that we have found a number of theoretical approaches
that fit with the provisional explanations. In some cases, these theories provide
more insight into the exact processes of the explanations whereas at the same
time, they give cause for further examination of a few variables and aspects we
Working irom Ajzen's theory, we also assume social influence
effects. As far as CRDs are concerned, the partner's influence as
well as the overt behaviour of befriended parents (modeling) will
undoubtedly be important (social support). In addition, we assume
that a number of people will be motivated (favourable attitude) but
probably will not be able to adopt the behaviour (low self- effica
cy). We will return to social influence and self-efficacy in the
course of additional research, when trying to find out what referen
ce persons are important and what the difficulties of performing
the behaviour are.
have not thought of yet. In practice, we would look for more theoretical
handles, and we would want to know whether theoretical constructs that look
promising were actually explanatory in our target population, doing additional
research.
For the development of planned health education programs via intervention mapping, it is necessary to understand the determinants of the target
behavior for the target population. Often, that understanding is lacking or
incomplete and additional research has to be done. In general, a combination of
qualitative and quantitative techniques is used to measure and analyze the
determinants of behavior (De Vries et al., 1992; see chapter 3). Usually theory
is used as the basis for framing research questions.
The first phase in measuring determinants involves a survey of the
available theoretical and empirical literature on the target behaviour or related
behaviors to find data and theories (which we have already done in the subject-
related approach). In the second phase, a qualitative method is used to find out
the target population's own ideas about determinants of their behaviour.
The third phase involves a quantitative method, a structured questionnaire with
questions that are based on the results of the qualitative phase and that is
administered to a large sample of the target population. During this process, the
16
General theories approach
At the same time, the application of Ajzen's model is an example
of a third possible strategy for association with theories, namely
the approach through general theories. If not through the con
struct-related approach, we would have thought of Ajzen's model
via the approach through general theories. We may apply another
general theory, McGuire's Persuasion-Communication Model
[McGuire, 1985], especially for the question about a possible
campaign failure. McGuire has distinguished a sequence of steps
from a first exposure to a health education message to the mainte
nance of the advocated behaviour: in this case, always using
CRDs. By means of McG^irejs model, we can give several reasons
why parents do not use CRDs (any more) or why the campaign
failed:
* attention: they have never heard of CRDs;
* comprehension: they do not understand the purpose of the CRD;
* attitude: they are not convinced of the advantages of the CRD;
* social support: the partner does not consider it necessary;
* self-efficacy: it is too much trouble when you have two cars;
* behavioural change: they do not think of it at the moment;
* behaviour maintenance: they tried, but do not like it. (Actually,
this last concept should also have come up in the construct-related
approach because one of the provisional explanations was that
parents stopped using CRDs.)
This list suggests that there may be various reasons why health
education has not produced the desired effect so far. Some of
these reasons have also been included in our provisional explanati
ons and have also come up via the other two approaches, but
some are new. In general, it seems that the determinants of CRDuse center around motivational issues and self-efficacy issues,
including practical barriers.
And again, it is useful to represent the theories one wants to apply in a
model consisting of boxes with arrows between them (Earp & Ennet,
1991). Some theories which seem to be clear and obvious at first sight
apparently are not so easy to be represented in a scheme.
15
fruitfully to the solution of the problem in question.
From the theoretical approach, our provisional explanations have been
improved and subsequently accentuated through additional information.
Sometimes the literature offers a wealth of empirical data. In this case
there was almost no information available, but we have been able to do
additional, quite elaborate research. In practice, this is not always possi
ble, and we will have to carry out additional research ourselves, often
with few time and means. Yet, we may achieve a lot with limited means,
for example by means of a number of focus group interviews.
Finding solutions
theories that he
come up in the earlier described protocol for applying
theories, will serve as a guideline for the literature search, the qualitative study
and the quantitative questionnaire. Be aware that some factors can not be
measures by just asking the target population; perceptions may be different
from realities, so we need information from key persons and through observati
ons.
Example - Applying Additional Research to the Problem of CRD Use
We now have a number of factors that are explanatory to our question.
Our task here is to summarize them and to tighten our provisional list into a final
list. These factors can also be thought of as indications for solutions. Two
criteria are used to select factors for solutions. One criterion is how important
the factor is. What is the strength of the relation between the factor and the
behavior in question. The second criterion that must be addressed now or later
in program development is how easy it is to influence a certain factor. In that
respect it is important to realize that some determinants may be changed by
interventions directed at the individual, but others by interventions directed at
the environment, often through changing the (decision) behavior of other actors
or organizations. Adequate solutions will center around factors that are
important and changeable. First we will brainstorm a list of provisional solutions.
Then we will again find theories through the issue-related approach, the
construct-related approach (constructs on the provisional list of solutions) and
the approach through general (change) theories. That will lead to a list of final
solutions that can be implemented and, preferably, evaluated.
Example - Finding Solutions for CRD Use
In the disappointing effects of the CRD-campaign in terms of the use of
18
In our example of CRDs, we will concentrate on the possibility of acqui
ring additional information through literature study and research. It may
be obvious that we do not know enough about the possible reasons for
not using CRDs. Even if we have listed all the reasons we do not know
which are the most important for our target population. This causes us
to do more research on the determinants of the acquisition and use of
CRDs or to appeal to existing research. In our example, we did a study
carried out by Pieterse et al. (1992). At the exit of a car park, Pieterse
and colleagues questioned parents and children about their reasons for
acquiring and using CRDs or not. Their study was set up according to
Fishbein & Ajzen's Theory of Reasoned Action, with specific attention for
practical barriers. In short, the researchers discovered that the safety of
the children (risk perception, attitude) is the main reason for acquiring
CRDs. More that 90% of parents were possitively disposed to use CRDs.
However, the most important reason for not using the CRD is the
child's response. If children become restless and have bad behavior
in a seat, parents often do not know how to cope with the behavi
or and consequently remove the child from the seat. In theoretical
terms, we speak of feedback about the negative consequences of
behaviour. These negative consequences result in low perceived
self-efficacy to continue the behaviour. In relation to the theories
regarding risk perception, we can see the possible impact of emp
hasizing the risks of not using CRDs. Emphasizing risk will have a
contrary effect on this group of parents due to their low perceived
self-efficacy to use CRDs. In this case high self-efficacy for using
CRDs would be an important requisite for effectively coping with
risk information. If we start thinking about solutions, the stress
should be on increasing self-efficacy rather than risk percepti
on.This case illustrates that a behavioral scientist not looking
farther than risk perception theories, cannot contribute much to
practice. The case represents an interesting application of risk
perception theory, but in this case, that theory does suggest the
wrong learning objectives and consequently does not contribute
17
who are supposed to adopt our solution. Again we are contrvnted with new
questions: what are the determinants for program implementers to adopt our
solution (question to the explanation) and how can we promote the adoption of
the program by these implementers (question to the solution). These questions
can be answered using exactly the same process as described earlier for
individual determinants and change. In practice, some theories are specifically
applied for implementation of interventions, such as diffusion theory (E.M.
Rogers, 1983), but in fact, adoption behavior by program implementers is
theoretically not different from individual behavior by the target population.
Determinants theories as the Theory of Planned Behavior can be applied to
understand adoption behavior (Paulussen et al, 1994; 1995) and change
theories as Social Cognitieve Theory can be applied to develop intervention
directed at program implementers (Parcel et al, xxxx).
I
non- use of CRDs we now have quite a lot of information. We know that
parents are already sufficiently aware of car seats so that an awareness
campaign will not be very effective. We also know that even though
parents tend to have some optimism bias in terms of their risk perception,
they do obtain and intend to use car seats. We know that a number of
factors are hard to influence, if at all. Regulation on child seats is not very
likely in the Netherlands, although it has been adopted in some countries.
A subsidy scheme does not seem feasible either. That is why these
factors are put aside for the time being. In addition, some factors are
easy to influence, but not so important, like making people aware of the
existence of child seats.
The most important factor to be influenced concerns the child becoming
restless if it is put in the seat. We know that this is an important reason
for parents not to use the acquired seat and at the same time, we
estimate that at this point, improvement may be achieved. We roughly
think of two kinds of solutions, namely a change in the environment:
improving the quality of seats, to make children feel more comfortable,
and a behavioral change: training and guiding parents in coping with
obstinate children. If we elaborate on that last solution, training parents in
coping, there are a number of theories that may help us developing an
adequate training program. Through the issue-related approach we could
not find any study on training parents for this particular situation. What
we did find were theories on learning I???] that would suggest that a
child who has ever been transported without a CRD presumably does not
want to be put in a seat any more. This means that parents have to
prevent children ever being transported without the CRD. Through the
construct-related approach we found theories on training, on coping, and
on the combination of those two, with Bandura's Social Cognitive Theory
as the most applicable one. Methods that may be used would be mode
ling with guided enactment, for instance in group meetings with parents.
Through the general theories approach we found theories on relapse
prevention (Marlatt & Gordon, 1985) that suggest ways to help parents
continue to use CRD's over time, by preparing them to deal with difficult
situations that would tempt them to return to non-use of the CRD.
The other possible solution, improvement of the quality of the CRD's, is a
different type of solution, in the sense that the intervention would not be
directed at the behavior of the parents or the child, but at the environ
ment. A change in the environment has to be organized systematically,
trough (the behavior of) various actors, such as parents, consumer
organizations, industry, and retailers. Theories that may help us develop
such an intervention are, for instance, theories on community develop
ment [XXX], coalition formation [XXX], and empowerment [XXX].
Having elaborated on our list of provisional solutions, we now can choose
solutions that are important and changeable. We will then continue and want to
implement the selected solutions, often working with program implementers
20
19
There are a number of over-all perspectives that authors
✓e used to
CHAPTER XX
HEORIES IN HEALTH EDUCATION
describe the determinants of behavior. Green & Kreuter (1991) distinguish
predisposing, enabling, and reinforcing factors. General social psychological
Objectives for chapter XXX: Find theories that are applicable to a problem
models, such as Theory of Planned Behavior [Ajzen, 1988] and Social Cognitive
at hand, choosing from a variety of theories of determinants and behavior
Theory [Bandura, 1986] distinguish basically three predisposing variables:
change
attitudes, perceived social influence and self-efficacy. We recognize these three
This chapter describes social science theories, primarily social psychologi-
determinants also in the Trans-Theoretical Model [Prochaska & DiClemente,
cal theories, that may be applied within the area of health education. Our list of
1984], but then as determinants of the progression through the stages. Finally
theories can only be a selection of all possible theories. This example can be
we see special health or risk related theories, such as the Health Belief Model
seen as a special case of the 'general theories approach'. We will describe four
[Janz & Becker, 1984].
broad categories of theories: theories on determinants of behavior to help with
Precede/Proceed
the choice of learning objectives, theories on behaviorxhange and environmen-
Green & Kreuter (1991) identify three categories of determinants of
tai changes tg help with the choice of methods, and theories on .implementation
behavior, individual as well as collective behavior, each of which has a different
to help with the anticipation of the implementation. For environmental changes
type of influence on behavior: Predisposing factors are those antecedents to
and implementation, the theories from the first two section are also applicable;
behavior that provide the rationale or motivation for the behavior: knowledge,
that is why we use the heading: 'additional' there. Table XX. 1 presents the
beliefs, attitudes, perceived social influence, self-efficacy.
Enabling factors are
theories we discuss here. Some of these are formal theories that have a
the antecedents to the behavior that enable a motivation to be realized: availabi-
standing tradition, others are very specific and sometimes represent only one or
lity, accessability, regulations, skills. Reinforcing factors are factors subsequent
more variables that are related to change. We refer readers to textbooks on
to a behavior that provide the continuing reward or incentive for the behavior
social psychological theories [Sabini, xxx; ????] or on health education theories
and contribute to its persistence or repetition: social support, peer pressure,
[Glanz et al, 1 996]. In Chapter 6 we will return to these theories, but then from
rewards & punishments.
the perspective of the search for appropriate methods.
In the following, we will elaborate on predisposing factors in our descrip
Table XX. 1: A laundry list of theories
tion of social psychological models. However, there is also a relation between
enabling factors and self-efficacy, and between reinforcing factors and percei-
22
Theories on Determinants of Behaviour
21
rably to an object, person, institution or event" [Ajzen, 1988, p.4]. Often, the
ved social influence. That is a complex relation, because PRECEDE/PROCEED
attitude is ('directly') measured by semantic evaluation, such as "good-bad".
also recognizes the difference between perceived social expectations (predispo-
TBP introduces the principle of correspondence, meaning 1 .at attitudes may
sing) and actual social expectations (reinforcing), as well as the difference
predict behavior when both are assessed at identical levels of target, action.
between perceived self-efficacy (predisposing) and actual skills (enabling). Over
context and time. An attitude towards a behavior is "the individual's positive or
time, people's initial behavior may be determined by their perceptions of social
negative evaluation of performing the particular behavior of interest" [Ajzen,
expectations and self-efficacy, but their ongoing behavior may be more effected
1988, p.117]. The attitude towards the behavior is determined by salient
by actual social expectations and skills. However, it is, for instance, possible
(behavioral) beliefs about that behavior. Each belief links the behavior to a
that people do not even try to change their behavior because of perceived lack
certain outcome or attribute ("Going on a low fat diet reduces my blood
of support, while in reality that support would have been available.
pressure"). Beliefs are weighted by the evaluations of those outcomes ("A
Social Psychological Models
reduced blood pressure is very good for me") and the 'indirect' attitude is the
Three major social psychological models for determinants of behavior are
summing of the multiplications of beliefs and evaluations.
Theory of Planned Behavior, Social Cognitive Theory, and the Trans-Theoretical
* Perceived social expectations. Ajzen and Fishbein use the concept 'subjective
Model [Eagly & Chaiken, 1993]. Ajzen's (1988) Theory of Planned Behavior
norms', which are also assumed to be a function of beliefs, but of a different
(TPB) is an extension of the Theory of Reasoned Action [Fishbein & Ajzen,
kind, namely the person's beliefs that specific, important, individuals or groups
1975]. In both theories, the first determinant of behavior is the intention to
approve or disapprove of performing the behavior ("Most people around me
perform that behavior (comparable to motivation as predisposing factor). TPB
think that I definitely should - definitely should not go on a low fat diet"). Asked
postulates that the intention is determined by three conceptually independent
for the different social referents ("My partner thinks..."), the beliefs are termed
determinants: attitude, subjective norms and perceived behavioral control.
normative beliefs and are weigted by the motivation to comply to referent
Ajzen's use of these concepts is somewhat confusing. We would prefer to use
persons or groups ("How much do you care what your partner thinks you
other names: perceived social expectations for Ajzen's subjective norms, and
should do?"). The 'indirect' perceived social expectations are the summing of
self-efficacy for Ajzen's perceived behavioral control. Ajzen (1991) indicates
the multiplications of normative beliefs and motivations to comply.
that his perceived behavioral control is not really different from Bandura's
Some authors distinguish between social expectations and social pressure, the
(1986) self-efficacy.
last being a much stronger influence [Evans et al, XXX; De Vries et al., 1995].
* Attitude. In TPB an attitude is "a disposition to respond favorably or unfavo-
24
23
"The types of outcomes people anticipate depend largely on the., judgments of
how well they will be able to perform in given situations" (p.392). So, when
people are not confident that they can use a condom consistently, they will also
* Perceived sei.
ficacy. Self-efficacy (Bandura, 1986) or perceived behavioral
control (Ajzen, 1988) refers to the subjective probability that one is capable of
executing a certain course of action. In TPB, this variable is measured by a
number of questions in terms of 'complete vs little control' or 'easy vs difficult'
not expect to prevent STDs.
* Observational learning. Most human behavior is learned by observation
("For me to go on a low fat diet would be easy - difficult"). Actually, Ajzen sees
through modeling. By observing others one forms rules of behavior, and on
self-efficacy not only as a determinant of the intention but also as a direct
future occasions this coded information serves as guides for action. Modeling is
determinant of the behavior, next to the intention. The idea is that self-efficacy
governed by four constituent processes:
* Attention for and perception of the relevant aspects of modeled activi-
has a relation with actual skills and barriers and therefore -at least partlypredicts actual behavior independent from the intention. Theoretically, this
presentation is rather confusing, but emperically the phenomenon is often
ties;
* Retention and representation of learned knowledge and rememberance;
supported.
Bandura's (1986) Social Cognitive Theory (SCT) covers both determi
* Production of appropriate action; and
* Motivation as a result of (observed) positive incentives.
Modeling is the influence of perceived behavior of (relevant) others, and should
be distinguished from TPB's perceived expectations of relevant others. Cialdini
et al. (1990) suggest two basic types of social influence: injunctive norms and
descriptive norms; injunctive norms are perceived expectations from others,
descriptive norms are perceived behavior of peers. Often, researchers would
assess both types of social influence in an extension of TPB: The attitude/socia!
influence/self-efficacy (ASE-)mode! of determinants of behavior [Kok et al.
1992?) [Schaalma et al., 1993][De Vries et al., 1995]. These three categories
of behavioral determinants can be seen as social cognitive perceptions, predisposing factors, which have to be distinguished from reinforcing factors (e.g.
26
nants of behavior and the process of behavior change. SCT explains human
behavior "in terms of a model of triadic reciprocality in which behavior, cogniti-
ve and other personal factors, and environmental events all operate as interac-
ting determinants of each other" (p.18). Major determinants of behavior in SCT
are outcome expectations, perceived self-efficacy, and observational learning.
* Outcome expectations and perceived self-efficacy. An outcome expectation
is a judgment of the likely consequence a certain behavior will produce ("When I
use a condom consistently, I will prevent STDs"). Perceived self-efficacy is a
judgment of one's capability to accomplish a certain level of performance ("I am
confident that I can use a condom consistently"). Outcome expectations are
comparable to behavioral beliefs in TPB. However, Bandura is very explicit about
the interrelation between outcome expectations and perceived self-efficacy:
25
Some researchers have encountered difficulties in applying the stages of
actual social support) and enabling factors (e.g. actual skills or barriers; [Geen &
change to behavior in case people do not recognize the risk (e.g. radio-active
Kreuter, 1991]. Ajzen (1988) and Bandura both (1986) call attention to the
radon: most people do not even know what it is), or do not think that the risk
potential discrepancy between perceptions of social norms and actual norms.
involves them personally. Questions about the intention to change are often
and between perceptions of self-efficacy and actual skills or barriers. Improving
meaningless for these respondents. Weinstein (1988) suggests an extension of
people's self-efficacy for healthy behavior through health education should be
the precontemplation stage: Stage 1: Has heard of hazard; Stage 2: Believes in
combined with lowering barriers that hinder healthy behavior through health
significant likelihood of risk for others; Stage 3: Aknowledges personal suscep-
promotion.
Prochaska & DiClemente's [Prochaska & DiClemente, 1984][Prochaska,
tibility.
De Vries & Backbier (1994) suggest that specific determinants are
DiClemente & Norcross, 1992]Trans-Theoretical Model (TMM) integrates several
involved in the different stage transitions: they present data on pregnant women
psychological constructs. A central construct in TMM are the Stages of Change:
and (non-)smoking as an example. To stimulate transition from precontemplation
people are thought to move from no motivation to change to internalisation of
to contemplation, people need to increase their perception of the benefits of
the new behavior. The early stages are defined by the intention to change the
changing the (problem) behavior and social support. In the contemplation stage
(problem) behavior while the later stages are defined by engaging in the new
an increase in perceived self-efficacy and social support with regard to engaging
behavior. The first stage is the precontemplation stage in which people have no
in the new behavior would result in transition to the action stage. To stimulate
intention to change their (problem) behavior. In a succesfull change process
transtion from action to maintenance, mainly an increase in perceived self
people transit to the contemplation stage in which they are thinking about
efficacy would be needed.
changing the (problem) behavior in the future (next six months). Than people
TTM can be used to describe, explain and predict behavior. As such, it is a
ideally move to the preparation stage in which they are planning to change this
model of behavioral determinants as well as a model of behavior change. Later
behavior on the short term (one month). People who have just changed the
on, in the part about theories of change, we will describe the theory again in
behavior are called actors, while people who have already internalized the new
terms of stages and processes of change.
behavior (for more than six months) are called maintainers. In the last one and a
Social psychological models of behavioral determinants do not imply a
half decade several versions of the Stages of Change are developed with
unidirectional influence; attitudes, social influence and self-efficacy can be
different definitions and stages. For example, earlier versions additionally
consequences as well as antecedents of behavior [Zimbardo & Leippe, 1991]
distinguished relapsers.
28
27
determined by his or her perceptions of personal susceptibility to, and the
Positive experiences with behavior may change psychosocial determinants of
seventy of, a particular condition of illness. The specific action taken is based
behavior, thus creating reciprocal determinism [Bandura, 1986] . (See Figure
upon a kind of cost-benefit analysis of perceived benefits and barriers. Accor
2.3).
ding to the HBM, this decision making process is triggered by a \ue to action'
which may be internal (i.e. symptoms of a disease) or external (e.g. health
education).
Health and risk related models. Historically, there have been a number of
theories that focus directly on health and risk related behavior [ Weinstein,
1988]. A model that has been used in a wide range of health related contexts
Although an impressive body of research findings has linked HBM
is the Health Belief Mode! (HBM) [Becker, 1974][Janz & Becker, 1984]. The
dimensions to health actions [Janz & Becker, 1984][Harrison et al., 1992],
basic components of the HBM are based upon psychological expectancy-value
recent research has demonstrated the importance of factors which were not
models hypothesizing that human behavior depends mainly upon the value
specifically developed or examined in the context of health behaviors. For
placed by an individual on a particular goal, and upon his or her estimate of the
example, many health-related behaviors are undertaken for reasons that are
likelihood that a given action will achieve that goal. With respect to health: the
ostensibly non-health reasons. suggesting that people's cost-benefit analysis
desire to avoid illness or to get well, and the belief that specific behavior will
should also include benefits other than health beliefs.
prevent or reduce illness. More specifically, the HBM consists of four psycholo
Current general social-
psychological models suggest, as we have seen, that an individual's behavior,
including health-related behaviors, is also determined by perceptions of social
gical variables [Janz & Becker, 1984]:
Perceived susceptibility, referring to one's subjective perception of the
influences, and by a conviction that he or she can successfully execute the
risk of contracting a particular condition or illness (perceived personal
behavior required to produce specific outcomes [Ajzen, 1988][Bandura, 1986].
risk);
Specific theories on determinants of behavior
* Perceived severity, referring to feelings concerning the seriousness of
Besides the general theories on determinants of behavior, there are a
contracting an illness;
number of specific theories that elaborate on one or more aspects of determi
* Perceived benefits, referring to beliefs regarding the effectiveness of
nants without claiming to be complete. Some of these variables are proposed as
various actions available in reducing the disease threat;
extensions of one of the general models and often can be seen as aspects -or
* Perceived barriers, referring to potential negative aspects of a particular
more distal determinants- of one of the three main determinants: attitude, social
health action.
influence, and self-efficacy.
In other words, an individual's decision to engage in a health action is
30
29
(including Ajzen, 1991) have shown that this variable can expiam extra variance
Risk pei^eption. As we have seen already, risk perception is a variable in
in behavior, next to current operationalizations of attitude, social influences and
a number of health and risk related theories, and it can be considered one
self-efficacy. Manstead & Parker suggest that the concept of personal norm
aspect of the attitude. Special theories explain the way people perceive risk.
could be related to anticipated regret [Richard et al., 1995]. When people try to
Unrealistic optimism [Van der Pligt, et al., 1993] is the tendency of people to
image how they would feel after having performed unhealthy behavior (for
think that they are invulnerable and that others are more likely to experience
instance unsafe sex), this anticipated regret stimulates future healthy behavior.
negative health consequences than oneself. There is considerable evidence that
The underlying assumption of anticipated regret is that people try to avoid
people are optimistic, however, the relation of this unrealistic optimism with
feeling regretful. Manstead & Parker suggest that anticipated regret could be
behavior is rather weak. There are cognitive and motivational causes for
seen as reflecting the anticipated affective consequences of breaking internali-
optimism:
zed moral rules, however most authors on anticipated regret suggest that the
* Cognitive: Perceived control ('I can control the risk'), lack of experience
concept is based in (negative) affect, not necessarily guild. The measurement of
(Never had an accident themselves or someone close), egocentric bias ('I
personal (moral) norms should comprise different operationalizations: "[the
have taken measures to prevent an accident'), sterotyped beliefs about
undesired (risky) behavior] would be wrong vs be right", "is appropriate vs not
people at risk ('Women get accidents').
appropriate for a person like me" [Godin & Kok, 1996], and "would make me
* Motivational: Self-esteem maintenance ('I am a better driver than most
feel guilty vs feel good" [Manstead & Parker, 1995]. Anticipated regret should
others'), defensive coping as one way to adapt to threats.
be measured by bipolar affective responses to the imagined undesired (risky)
Perceived personal risk is often a necessary condition for change, but it is
behavior in terms of "would make me satisfied vs dissatisfied" and "would
seldom a sufficient condition for change.
make me anxious vs not anxious" [Richard et al, xxx]. For both variables the
responses towards the desired healthy behavior could also be measured.
Attribution theories. An important variable in many models that try to
Persona! (moral) norms and Anticipated regret. A number of researchers
have suggested that the Theory of Planned Behavior (TPB) should be extended
with a variable personal norms, or personal normative beliefs, moral norms, self
explain determinants of behavior is self-efficacy. An interesting question is:
or role-identity [Manstead & Parker, 1995][ Godin & Kok, 1996]. Personal
What are the determinants of self-efficacy? Weiner (1986) suggests that self
norms are beliefs about what is right and what is wrong to do. In this meaning,
efficacy (Weiner: 'expectancy of success' but we prefer 'self-efficacy') is
personal norms are not an aspect of social expectations or subjective norms but
determined by the perceived stability of the attributions for success and failure.
are probably part of the attitude. Using different operationalizations, researchers
32
31
this means that we have to find the high-risk situations that peoH>e are not able
A person attributing a failure to a stable cause (e.g. ability) will have a lower
to cope with. Most relapsers are quite able to indicate these situations. Measu-
self-efficacy for performing the same task again, compared to somebody who
res of self-efficacy should, among others, be operationalized as estimations of
attributes a failure on the same task to an unstable cause (e.g. luck). After
confidence to cope with various difficult (high-risk) situations.
success this effect is reversed. Furthermore, attribution theory assumes that a
A summary of determinants of behavior
lower self-effiacy leads to a less adaptive task behavior; people will invest less
Summarizing the theoretical ideas on determinants of behavior (and
maintenance), we now present a model of determinants.
Figure XX. 1
In Figure XX. 1, the determinants are visually organized. The third column
energy in the task at hand. Support for an attribution explanation of health
behavior is found, among others, in a study by Hospers and colleagues (1990).
They show that the success of participants in a weight reduction program was
positively related to their self-efficacy at the start of the program. Self-efficacy
represents Green & Kreuter's (1991) ideas, the second column represents an
was negatively related to stability of attributions for earlier failures, and both
extended version of Ajzen's (1991) Theory of Planned Behavior, adapted with
relationships were independent of the number of failures.
Bandura's (1986) ideas: the ASE-model. It is important to recognize the feed
TTM and attribution theories both recognize that by using theories on
back loop: determinants and behavior have a reciprocal relationship. Moreover,
determinants of behavior, we sometimes try to explain the behavior of relap-
the model is open at the left side of every box; there are many other variables -
sers: People that have tried to change their behavior, but failed. Basically,
for instance: personality characteristics, contextual factors- that influence
relapse prevention theories are theories for health behavior change, but again,
behavior, through these determinants of behavior, but the basic idea is that the
they may also help in understanding current determinants of behavior. To
model represent the major intermediate determinants. Previous behavior is
understand the determinants of smoking in individuals that repeatedly failed
sometimes mentioned as a determinant of future behavior. That is basically the
quiting smoking, we have to know why they failed and how they attribute these
same as predicting tomorrow's weather from today's: very often the prediction
failures (see Attribution theories). A key concept in Relapse Prevention Theory
is right, but it does not give any insight in the underlying process [Ajzen,
[Marlatt & Gordon, 1985] is the so-called high-risk situation. A high-risk
1991][De Vries et al., 1995].
situation is a situation in which people are tempted to return to their former
The arrows from Reinforcing factors to Perceived social influence, and
(unhealthy) habits. In order to cope with high-risk situations, people need
from Enabling factors to Perceived self-efficacy indicate a indirect influence next
adequate coping responses. Relapsers obviously do not have sufficient coping
to the direct influence. For instance: Barriers have a direct negative influence on
responses, resulting in low self-efficacy and relapse. For determinants analyses,
34
33
Theories on behaviour change through communication
Current general models on behaviour change distinguish steps, phases or
Behavior, tx
«so an indirect influence through Perceived self-efficacy and
Intention on Behavior. The direct relation between Perceived self-efficacy and
stages of change, especially McGuire's Persuasion-Communication matrix,
Behavior is an empirical phenomenon in TPB research, that Ajzen (1988)
Prochaska & DiClementi's Trans-Theoretical Model, and Rogers' Diffusion
explains by assuming that Perceived self-efficacy is a better reflection of the
Theory. The major determinants theories, Bandura's Social Cognitive Theory,
direct relation between Enabling factors and Behavior, than Intention. In the
Ajzen Theory of Planned Behavior, and Green & Kreuter's Predisposing, Enabling
model, the vertical relations between the determinants are left out; these may,
and Reinforcing factors, can also be seen as change theories, in the sense that
however, be strong or weak, depending on the behavior, the person and the
they indicate which determinants have to be changed, and, especially in the
situation. Our knowledge is as yet limited as to the conditions under which
case of Social Cognitive Theory: how those changes may be brought about.
certain determinants are more influential than others (Ajzen, 1988, p. 138-142;
One general framework for theories on behaviour change is provided by
McGuire, 1991). We do not present the model in figure 1 as a formal theory.
McGuire s (1985) Persuasion-Communication Model. This model describes the
but as a working model that identifies the major concepts that may function as
various steps that people take, from the initial response to an educational
determinants of behavior, and that reflects the interrelations among those
message to, hopefully, a continuous change of behaviour in the desired directi-
on. Simplified, the first steps refers to successful communication, the subsequent steps refer to changes in attitudes and behaviour, and the last step refers
to the maintenance of that behaviour change. Going through these steps.
McGuire argues that the educational interventions should change with each
step. The choices that have to be made about the message, the target group,
the channel, and the source, may be different or conflicting, depending on the
concepts. We are aware of shortcomings in the working model, in terms of
preciseness of concepts and relations. However, we do not think that any
theory is comprehensive enough to cover all the different theoretical ideas.
Other authors have tried to produce a comprehensive working model of theories; Flay [xxx], for instance, suggests a comprehensive model that is compara-
ble to what we present here. Another interesting approach is the so-called
behavior mapping process that ??? [???] recently developed; starting from
predisposing, enabling and reinforcing factors, they draw a map that shows all
particular step that is addressed.
Prochaska & DiClemente's Trans-Theoretica!Mode! (TTM) distinguishes
the so-called Stages of Change within the person: pre-contemplation, contem-
plation, preparing for action, action, and maintenance or relapse [Prochaska &
DiClementel 984][Prochaska et al., 1994]. Their model does not refer to the
the possible routes into specific theories (in their case, 'mapping' is a visualizati-
on, not a process). We suggest that a multi-theory approach is the best strategy
when applying behavioral science to health education questions; a problem-
driven applied behavioral science approach [Glanz, 1995][Kok et al., 1996].
35
36
V-",
communication process, but the similarities between this model and McGuire's
influence behaviour.
As described earlier, the Theory of Planned Behavior distinguishes
attitudes, perceived social expectations and self-efficacy. Green & Kreuter
distinguish Predisposing, Enabling and Reinforcing factors. Combining TPB and
SCT, we distinguished three types of (predisposing) dterminants: Attitudes/Out-
comes, Perceived Social Influence, and Self-Efficacy Expectations (see figure
model are evident (see figure XX.2). An important contribution of the stages-of-
change model is the specific tailoring of educational efforts to groups of people
in different stages of change. Interventions based on this model normally have
completely different methods or strategies for each stage. Another construct in
TTM are the Processes of Change-, the experential and behavioral processes.
These processes are thought to mediate the cognitive changes in the next two
XX.1).
constructs in the TTM: The Decisional Balance, referring to the pros and cons of
Comparing and combining all the various general change theories, we see many
constructs that are basically similar, see figure XX.2. We can see that most
theories assume some kind of order in a series of changes: steps, stages,
phases. We will continue from here on with six steps from figure XX.2: 1)
successful communication, 2) attitude change, 3) social influence change, 4)
self-efficacy change, 5) behavior change, and 6) maintenance of behavior
change. Intention or decision is not one of the steps, because we assume that a
change of intention follows directly a change in attitude, social influence and
self-efficacy. As with the model in figure XX. 1, we see this six steps framework
as a working model, not as a formalized theory.
Within this general framework, a number of other theories can be applied
[Zimbardo & Leippe, 1991][Glanz et al., 1990][McGuire, 1985; 1991]. Alt-
hough these theories often cover only steps, or even only parts of steps, they
can be helpful in developing interventions that focus on particular aspects of
change. Using McGuire's framework, in turn, can be helpful in stimulating
program planners to recognize neglected variables and to recognize appropriate
38
the (problem) behavior and Temptations to Engage in the (problem) behavior
which is related to self-efficacy.
E.M. Rogers' [1993] Diffusion of Innovations Theory will be described in
detail later in this chapter, and in Chapter 6. The diffusion theory is often seen
as a theory for implementation, but it is also applicable for individual changes.
Rogers distinguishes four steps: dissemination or awareness, adoption or
decison, implementation or action and continuation or institutionalisation.
One general theory, or theoretical framework, covering both determinants of
behaviour and the process of behaviour change is Bandura's (1986) Social
Cognitive Theory (SCT). In SCT the relationships between cognitive, environ-
mental and behavioural variables are seen as interactive and bi-directional.
Reinforcement of behaviour is a key environmental factor studied by social
cognitive theorists. Other people in the environment can also affect behaviour
because a person learns through observing others and receiving reinforcement.
The SCT cognitive variables include outcome expectations and self-efficacy
expectations. Modeling and incentives are SCTs major intervention methods to
37
2,
theories that can be applied.
age, ethnic groups, differences in health education and experience etc.
Interest in a message is also concerned with personal risk perception
Successful communication
[Weinstein, 1989][van der Pligt, 1991]. According to Weinstein, people tend to
Succesfull communication is often operationalized as exposure, attention
unrealistic optimism: they systematically underestimate their own personal risk
compared to the risks of others. It is owing to unrealistic optimism that they
often fail to take enough measures to protect themselves against a given risk.
The principal causes of this optimism are probably that they underestimate what
others undertake to protect themselves and that they conceive stereotypes of
effect only if the target group is exposed to it and pays attention to it, which
may involve processes of selective exposure and selective perception [McGuire,
1985]. These processes depend on, among other things, the situation people
find themselves in. A student of a strictly religious school is relatively less likely
people running high risks.
A health education message has to be clear and may be repeated several
I
I
and comprehension (McGuire, 1985). A health education message will have
to be exposed to information about condom use. At the same time, it is a
times [McGuire, 1985]. However, not all receivers are equally interested in the
matter of motivation: the same student is probably less interested in information
message. In this respect as well as in the next step (attitude change), the
about condom use. In the first case, it already helps if health education is
Elaboration Likelihood Model of Petty & Cacioppo (1986) has been found a
provided at school. In the second case, this will not be enough and the student
relevant theory. Some people have a strong 'tendency to think', i.e. they tend
should be exposed to extra motivational messages, for example by choosing
to think carefully about the message arguments (central route processing).
interesting materials which are compatible with the student's life style. Compa
Others show less tendency to think and are more responsive to peripheral cues
tibility usually is not as easy as it may seem [E. Rogers, 1983]. In general,
such as the source, the form of the message and the behaviour of others
health educators are highly educated, enjoy a high socioeconomic status, have a
(peripheral route processing). Changes effected through the central route are
thorough command of the current language, etcetera. Most target groups are
likely to persist longer than changes affected through the peripheral route.
totally different. In theory that should not be a problem, but even professional
Health education should try to encourage central processing both by motivating
health educators tend to underestimate this difference. Health education has to
receivers to think and handing skills for careful information processing [Petty &
be carefully geared to target groups in terms of speech, priorities, the credibility
and attractiveness of the source and the message, the clearness of the messa-
Cacioppo, 1986].
ge, compatibility with standards, values and experience etc. In addition, the
Attitude change
People may be aware of the severity of a problem such as cardio-
40
health educator has to reckon with potential subgroups: differences in sex and
39
vascular diseases, cancer and aids, thinking that they will not be affected by
motivating [Eagly & Chaiken, 1993][Tversky & Kahneman, 1986].
such diseases themselves. We then say that they are aware of the severity of
The so-called health beliefs (fear appeal) sometimes wrongfully lead to
the fact that these health reasons are more attended to than other variables
■
the problem but not of their own susceptibility. These two concepts are part of
when examining the attitude towards prevention behaviour. However, outcome
theories about risk perception and fear- arousing communication: Health Belief
expectations cover more than mere health expectations, as we have already
Model [Janz & Becker, 1984], Protection-Motivation Model [R. Rogers, 1983-
stated when discussing models such as Health Belief Model and common
]and Emotion Model [Levental, 1984]. These theories show that fear and threat
models about attitudes being one of the behavioural determinants [Fishbein &
caused by severity and susceptibility may incite people to action. However, the
Ajzen, 1975].
concrete form of this action strongly depends on the so-called outcome expec
An attitude is formed by balancing all relevant advantages and
disadvantages (outcomes) of both the desirable and the undesirable behaviour.
tations [Bandura, 1986]: 'Would it help if I did this?' and perceived self-efficacy:
Moreover, people generally are susceptible to short-term considerations rather
'Can I manage to do this?'. A tendency exists, especially among non health
than long-term considerations [McGuire, 1985]. Health education always has to
educators, to turn to increasingly though and thus fear-arousing communication
be based on the analysis of the determinants of behaviour. Nevertheless, we
in case of serious problems. It can, however, be stated on the basis of the
should realize that people nearly always notice important short-term advantages
above theories as well as available empirical data that fear can be a bad
of the undesirable behaviour as well. To illustrate, McGuire (1991) outlines the
counsellor. Health education about cardio-vascular diseases, cancer and aids
reasons for adolescents to use drugs: resistance to status quo, sensation
usually is fear-arousing enough as it is, and the health educator should rather
seeking (taking risks), impression management (maturity) and joining specific
attempt to persuade people of the use of prevention (outcome expectation) and
to increase their perceived self- efficacy of prevention behaviour (see section.
subcultures. Similar reasons may apply to unsafe sex.
Increased self- efficacy). High-fear appeal in combination with low self-efficacy
Health education often implies that health educators and receivers take
different positions at first instance. For example, a health educator thinks that
adolescents should always wear condoms whereas the adolescent himself
I
L
HIV-infection or searching for scapegoats such as homosexuals or drug addicts.
This has happened in countries where aids education has been fear-arousing
considers this superfluous. We then speak of discrepancy between source and
message on the one hand and source and receiver on the other. Discrepancy is
can lead to dysfunctional behaviour, for example, the denial of personal risk for
i
without giving clear and feasible advice regarding prevention [Winn, 1991]. If
one of the variables in the matrix of McGuire which affects changes differently,
clear and feasible advice is provided in order to remove threat, referring to the
sometimes even adversely, in various steps. [Fishbein & Ajzen, 1975, p.469]. In
loss people will suffer if they do not follow this advice apparently can be very
42
41
people. This does not constitute a problem in educational s:+uations involving
terms of possible changes, we can state that the higher the discrepancy, the
individual contacts, though it does in more mass media health education where
greater the potential change. However, in terms of acceptance, the opposite is
the message should be formed so as to fit in with the majority of the receivers,
true: the higher the discrepancy, the smaller the chance of acceptance and thus
which is hard to realize in case of considerable individual differences.
of change. This implies a curvilinear, reversed U-shaped relation between
Finally, emotional or affective aspects are very important to attitude
discrepancy and change: first, at intermediate levels of discrepancy, there is
development and attitude change. It is essential that people's first emotional
more change when discrepancy is high. However, when discrepancy is beyond
response to health education is positive. This can be achieved by, among other
a certain level, less change occurs when discrepancy increases. In addition,
things, repeated exposure and association with other stimuli which have already
other variables play a part in it, such as ego-involvement and the credibility of
caused a positive emotional response [Zimbardo & Leippe, 1991],
the source. Ego-involvement implies the extent to which a receiver is emotional-
’
Social influence change
ly involved with the subject of a message. For example, adolescents who are
Social Comparison Theory [Suls & Wills] argues thct people like to be
seropositive or know seropositives in their direct environment are more involved
equal to other people as far as opinions are concerned and slightly better as it
with class room discussions about the stigmatizing of seropositives than
comes to abilities. Yet the others people must be referents, i.e. other points of
adolescents who do not know any seropositives. The more ego- involved a
similarities are required between the receiver and the others. An adolescent
person is with the issue, the less likely the change through a discrepant messa-
having much experience with relations and sex is not very likely to compare
ge. The credibility of the source is a combination of expertise, integrity and
himself with an age-mate in an educational video who presents himself as an
attractiveness. The higher the credibility, the higher the chance of change.
advocate of abstinence outside the marriage. People tend to divide themselves
Especially if the source comprises an unanimous group of referents (see section:
and others into categories, assimilating with groups they (would like to) belong
social influence), it may have a strong influence. This whole of influencing
to and contrasting with groups they do not (want to be) a member of [Turner,
effects creates a complex situation which is hard to simplify [Fishbein & Ajzen,
1991]. The same phenomenon also explains why prejudices and discrimination
1975]. Health educators can try not to be too discrepant by, apart from ego-
still exist. Social comparison is important, particularly if objective information is
involvement, stressing personal interest in behavioural change (response
scarce or lacking. In such cases, people tend to conform themselves to referen-
involvement) and by being a reliable source. Here a problem occurs which we
ce others ('conformity', [Festinger, 1954] 'social norms', [Fishbein & Ajzen,
have already mentioned before (among others, when discussing tailoring),
1975], 'social pressure' [Evans et al, ???]), especially if this reference group is
namely that a receiver variable like discrepancy may strongly differ between
44
s
43
|
T
1
Le>
\
By enacting the behaviour, they learn by the impact of the behaviour. Most
behaviour is learned by a combination of observing and enacting, supported by
feedback and reward. Bandura (1986, p.161) argues that 'modeling with guided
enactment' is the most optimal method to increase self-efficacy and modify
behavior. Under the next two paragraphs we will describe other methods for
self-efficacy improvement that are linked with behavior change and maintenan
unanimous. The fact that these others provide information about social reality is
part of the explanation, as is the fact that they offer social reward. For example,
an adolescent joining a couple of friends on a holiday for the first time who
notices they all take condoms with them will be tempted to the same because it
seems to be wise and it feels good to be one of them.
On the other hand, conformity often forms an obstruction to behavioural
change. A student who intended to use condoms actually may be restrained
ce.
from it due to the opinions and behaviour of others. Research on conditions
Behavioural change
The change from increased self-efficacy to behavioural change is not
marked by a sharp line. Some of the principle variables in these two steps are
equal. This is already imbedded in the concept 'enactive learning'. Providing
positive feedback constitutes a great problem in many health education cam
paigns [Strecher et al., 1995]. Although in the long run, the outcomes of the
advocated behaviour usually are profitable, hardly any positive effect can be
experienced on short notice. However, positive feedback is an important reward
which gives people an incentive to overt behaviour and maintenance (see
section: behavioural maintenance). That is why health educators should strive
under which the tendency towards conformity decreases [Turner, 1991] shows
that the violation of unanimity is very effective: as soon as a reference group
includes one ally, conformity immediately decreases. Conformity will also be
reduced by social influence of an ally or reference group that happens to be
absent at that moment. Health educators attempting to make people resist
pressure often use strategies based on increasing resistance to pressure.
Moscovici (1985) argues in his theory about minority influence that a minority
may influence the behaviour of a majority by showing their own consistent
behaviour, without becoming rigid and still being part of the group. All these
resistance and forewarning techniques subscribe the necessity for the target
for positive feedback.
Feedback is also an important concept in Locke's theory about goal
setting [Locke & Latham, 1991 HStrecher et al., 1995]. Locke has demonstrated
that setting a challenging goal, i.e. difficult though feasible, leads to a better
performance than setting an easy goal or no goal at all. This positive effect of
defiant goals occurs if a person disposes of sufficient experience, self-efficacy
and feedback, and accepts the challen- ge. Goal setting leads to better perfor-
group to learn the skills needed to resist that pressure because without impro
ving self- efficacy, forewarning may be counterproductive.
Increased self-efficacy
Bandura's (1986) Social Cognitive Theory regards behavioural change as
a kind of learning, namely the result of observational learning and enactive
learning. By observing others, people learn skills and new behavioural patterns.
45
46
■
.
[Weiner, 1983] and relapse prevention theories [Marlatt & Gordon, 1985].
Behavioural maintenance is essential to health education; behavioural change
makes sense only if it is continued. It happens quite often that people showing
new behaviour receive negative feedback or find themselves in so-called high
mances because people exert themselves more, persevere in their tasks,
concentrate more
and if necessary, develop strategies. As far as aids prevention
is concerned, the health educator may attempt to associate safe sex with
important goals of students such as their careers which might be threatened by
risk situations, thus creating the risk of relapse. People who have relapsed into
the possible consequences of safe sex. In this way, safe sex becomes part of
their former behaviour several times and have attributed this to stable causes,
the strategy to attain long- term objectives.
will develop low self-efficacy and feel helpless. They should learn that they
failed due to instable causes, and that the advocated behaviour requires abilities
which they are able to acquire. Relapse prevention techniques teach people how
to deal with high risk situations. In practice, techniques to increase self-efficacy
and enhance behavioural change are also used in coping with negative feed
back, high risk situations and the promotion of behavioural maintenance. People
of the same target group may find themselves in different stages of change and
therefore, it is important to continue health education in one way or the other,
for example, by means of boosters (Flay et al., 1989). Such boosters may dilate
upon problems encountered or provide materials which later will meet the need
In educational campaigns aimed at behavioural change in particular, so-
called commitment techniques can be used [Kiesler, 1971], i.e. clearly visual
positions such as public commitment or
the public demonstration of an advoca
ted behaviour. The latter may consist of, for example, young people participa
ting in an educational video about aids prevention. The effect of commitment
can be explained partly by cognitive dissonance [Aronson, 1991], which causes
people to adapt their opinions to their behaviour.
Zimbardo & Leippe (1991) have argued that actual prevention behaviour
may be motivated by so-called prompts, i.e. recollections of one's intentions at
the right time and the right place. Such a prompt ensures that good intentions
are made salient when needed. Health educators may help their target group to
then prevailing.
find the most appropriate place for these prompts. In doing so, they may apply
Additional theories on environmental change
Interventions targeted at determinants of behavior (or even directly at
determinants of the problem) may be located inside the individual such as
for the so-called anticipated regret [Richard et al., 1991], making people imagine
how they
would feel after performing the undesirable behaviour. Such a
strategy has proved capable of keeping people from the undesirable behaviour.
attitudes, perceived social influence and self-efficacy, but they may also be
Behavioural maintenance
located outside the person in the environment, the physical environment as well
Recently, health educators have been more involved in behavioural
as the social environment. In practice most determinants are combinations of
48
v ■
maintenance [den Boer et al., 19911 than in the past, using attribution theories
47
4
promotion planning process. If we do not ensure implementation, our work has
personal and environmental determinants. Self-efficacy (personal) is related to
been largely wasted. School programs for the prevention of smoking are useless
barriers (environment); perceived social expectations (personal) are related to
if teachers do not use them. Underestimating diffusion ar J adoption barriers is
real social expectations (environmental). The distinction between personal and
one of the reasons for health education being sometimes ineffective. While the
environmental for determinants is not necessarily the same as for interventions:
need for information on determinants of individual behaviour is commonly
A personal determinant may be intervened upon through personal as well as
accepted, we often fail to recognize that, to develop implementation strategies.
environmental interventions, for instance: in case of social pressure for unheal-
we also need information on determinants of institutional 'behaviour', such as
thy behavior, the intervention could focus on resistance to social pressure
adoption of a prevention program by organizations or decision-makers within
(personal), but also on changing social influence through group methods
those organizations. As we mentioned earlier, all theories on individual determi-
(environmental). Comparably, interventions could focus on self-efficacy impro
nants and individual behavior change may be appied for implementation behavi-
vement but also on reducing barriers. Of course, an optimal intervention would
or. There are some special theories that we see often applied for implementati-
try to combine both types of interventions.
on, but they could also be useful in individual behavior. The existing knowledge
Very often, environmental changes are brought about by changes in other
in theories on the diffusion and adoption of health promotion will be summarized
people's behavior or organizational behavior: gouvernmental decision makers,
in Chapter 6. Here we will just mention three perspectives: features of the
managers, elections, school boards, hospital boards, etcetera. Basically, these
innovation that determine adoption, program implementers' behavior, and the
behavior changes follow the same course as individual behavior change, as we
importance of a linkage system.
described earlier. There are a number of theories, however, that focus specifi-
Classical research in the area of diffusion and adoption, mostly in
cally on influencing the environment. We will describe them shortly.
schools and worksites, suggest a number of features of an innovation (the
Community development ???
health education intervention) that determine (non)adoption [Orlandi et al.,
Social action ???
1991]. These are : Compatibility, Flexibility, Reversibility, Relative advantage.
Coalition formation ???
Complexity, Cost-efficiency, Risk.
Empowerment ???
The focus of contemporary research on
diffusion of health interventions gradually shifts from features of an innovation
to program implementers' planning behavior and thought processes with regard
to awareness, adoption, implementation, and continuation of innovations.
50
Additional theories on implementation
Implementation of a prevention program is an essential part of the health
49
i
Bartholomew et al, 1990;
Bartholomew et al. 1991]
Within this decision making approach, the abovementioned features of innovations can be dealt with as subjective expectancies about advantages and disad-
Becker, M. (1974). The Health Belief Model and sick role behavior. Health
Education Monographs, 2, 409-419.
I
Cooper H Hedges LV eds The handbook of research synthesis. New York
Russell Sage Foundation, 1 994.
De Vries, H., Backbier, E., Kok, G., & Dijkstra, M. (1995). The impact of social
influences in the context of attitude, self-efficacy, intention and previous
behaviour as predictors of smoking onset. Journal of Applied Social Psychology,
25, 237-257.
With regard to intervention development and the anticipation of factors
I
that may impede or improve intervention diffusion, Orlandi and colleagues stress
the need for a linkage system between the resource system that promotes the
intervention (e.g. the Anti-Cancer Council) and the user system that is supposed
to adopt the intervention (e.g. schools). Such a linkage system should include
1
Eagly, A.H., Chaiken, S. (1993). The psychology of attitudes. Orlando: Harcourt
Brace, Jovanovich.
The theory of planned behavior can be applied to implementation behavior as
well [Paulussen et al, 1 994][Paulussen et al., 1995].
De Vries, H., Weijts, W., Dijkstra, M., & Kok, G. (1992). The utilization of
qualitative an quantitative data for health education program planning, imple
mentation and evaluation: a spiral approach. Health Education Quarterly, 19,
101-115.
den Boer et al., 1991]
vantages of innovation adoption, implementation, and continuation respectively.
representatives of the user system, representatives of the resource system, and
a change agent facilitating the collaboration. Again, the theories that were
mentioned in the paragraph on interventions can be applied here to develop
Earp & Ennet, 1991
interventions for adoption and implementation.
Evans and colleagues (1981)
Epilogue
Festinger, 1954
STILL HAS TO BE WRITTEN
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention and behavior: An
introduction to theory and research. Reading, MA: Addison Wesley.
Glanz, K. & Rimer, B., 1995. Theory at a glance; a guide for health promotion
practice. U.S. Department of Health and Human Services, National Institutes of
Health/National Cancer Institute, NHI Publication No. 95-3896.
References
Godin, G. & Kok, G., 1996. The theory of planned behavior: A review of its
applications to health-related behaviors. Under review.
Ajzen, I. (1988). Attitudes, personality, and behaviour. Milton Keynes, UK:
Open University Press.p. 138-142
Green, L.W., & Kreuter, M.W. (1991). Health promotion planning; an educatio
nal and environmental approach. Mountain View, Cal.: Mayfield.
Aronson, 1991
Bandura, A. (1986). Social foundations of thought and action. Englewood Cliffs,
NJ: Prentice-Hall.
Green & Lewis, 1986
[
Greenbaum TL 1988 The Practical Hanbook and Guide to Focus Group Re
search. Lexington, MA Lexington Books.
52
Bartholomew et al. 1988;
51
Harrison, J.A., Mullen, P.D., & Green, L.W. (1992). A meta-analysis of studies
of the Health Belief Model with adults. Health Education Research, 7, 107-116.
Messick & Brewer, 1983
McGuire, W.J. (1991). Using guiding-idea theories of the person to develop
educational campaigns against drug abuse and other health-threatening behavi
our. Health Education Research, 6, 173-184.
Hedges LV Olkin (1985)Statistical methods for meta-analysis. New York:
Academic Press,
Hospers, H.J., Kok, G. & Strecher, V.J. (1990) Attributions for previous
failures and subsequent outcomes in a weight reduction program. Health
McGuire, 1985;
Merrrian, SB (1988) Case Study Research in Education: A Qualitative Appro
ach. San Francisco: Jossey Bass.
Miles, M. (1979) Qualitative data as an attractive nuisance: The problem of
analysis. Administrative Science Quarterly, 24, 590-601.
Miles M & Huberman, M. (1984). Qualitive data analysis: A source book of new
methods. Beverly Hills, Ca: Sage.
Morgan DL (1988) Focus Groups as Qualitative Research. Newbury ParkSage
Publications
Moscovici (1985)
Education Quarterly, ]!_, 409-415.
Huberman AM & Miles MB (1994). Data management and analysis methods. In
N.K.Denzin & Y.S. Lincoln (Eds.) , Handbook of qualitative research (pp. 428444). Thousand Oaks, CA: Sage.
Janis & Mann, 1 977
Janz, N.K., & Becker, M.H. (1984). The Health Belief Model; a decade later.
Health Education Quarterly, 1J_, 1 -47.
Jones et al., 1 984
Katz, 1981
Mullen PD Ramirez G (1987) INformation Synthesis and Meta Analysis in
Advances in Health Edcuation and Promotion, Vol. 2, Pages 201-239.
Kiesler, 1971
Orlandi et al., 1991
Kok, G., Schaalma, H., De Vries, H., Parcel, G. & Paulussen, Th., 1996. Social
psychology and health education. In: W. Stroebe & M. Hewstone (Eds.),
European review of social psychology, Volume 7, in print.
Parcel et al., 1989a
Koomen, 1 992a
Parcel et al., 1989b
Kreuger RA (1988) Focus Groups A Practical Guide for Applied Research.
Patton, 1990
Newbury Park CA Sage Publications
Paulussen et al, 1994
Lave & March (1975),
Paulussen et al., 1995
Lazarus, 1991
Petty & Cacioppo (1986)
Levental, 1 984]
Prochaska, J.O., & DiClemente, C.C. (1984). The transtheoretical approach:
Crossing traditional boundaries of therapy. Homewood, III.: Dow Jones-Irwin.
Locke & Latham, 1991]
Prochaska, J.O., DiClemente, C.C. & Norcross, J.C., 1992. In search of how
people change; applications to addictive behaviors. American Psychologist, 47,
Manstead, A.S.R., & Parker D. (1995). Evaluating and extending the Theory of
Planned Behaviour. In W. Stroebe, & M. Hewstone (Eds.), European Review of
Social Psychology, vol. 6 (pp. 69-96). Chistester, UK: John Wiley & Sons.
1102-1114.
Richard, R., van der Pligt, J., & de Vries, N. (1995). Anticipated affect and
behavioural choice. British Journal of Social Psychology, in press.
54
Marlatt, G.A., & Gordon, J.R. (1986). Relapse prevention; maintenance straje^
gies in the treatment of addictive behaviours. New York: Guilford.
53
‘■f,y
Rogers, E, 1990
Table 2.4: A laundry list of theories
Variables:
Rogers R W, 1983
Theory/Author(s):
Predisposing/Enabling/Reinforcing
Attitude/Subjective norm/
Perceived behavioral control
Outcomes/Modeling/Self-Efficacy
Social pressure
Stages of change
Health beliefs/Benefits/Barriers
Personal risk/Unrealistic optimism
Personal moral norm
Anticipated regret
Attributions/Success expectancies
Coping with high-risk situations
Rossi PH FreemanHE, 1989 Evaluation: A Systematic Approach. Newbury
Theories on determinants
PRECEDE/PROCEED, Green & Kreuter
Planned Behavior, Ajzen
Park: Sage
Schaalma, H., Kok, G., & Peters, L. (1993). Determinants of consistent condom
use by adolescents: The impact of experience of sexual intercourse. Health
Social Cognitive theory, Bandura
Evans
Transtheoretical model, Prochaska & DiClemente
Health Belief Model
Precaution Adoption Model, Weinstein
Godin
Richard
Attributional theories, Weiner
Relapse prevention, Marlatt & Gordon
Theories on behavior change
Persuasion-Communication, McGuire
Successful communication/Changes in
determinants & behavior/Maintenance
Stages of change/Processes of change
Transtheoretical model, Prochaska & DiClemente
Social Cognitive Theory, Bandura
Modeling/Active learning/lncentives
Selective exposure/Selective attention
McGuire
Compatibility
Diffusion Theory, E.M. Rogers
Personal risk/Unrealistic optimism
Precaution Adoption Model, Weinstein
Central route/Peripheral route
Elaboration Likelihood Model, Petty & Cacioppo
Fear/Danger/Outcomes/Self-efficacy
Protection Motivation, R. Rogers
Reasoned Action, Fishbein
Outcomes/Advantages and disadvantages
Planned behavior, Ajzen
Short term outcomes/Long term outcomes
McGuire
Discrepancy/lnvolvement
McGuire
Emotional response
Zimbardo
Social comparison
Social comparison, Suls
Conformity/Social pressure
Festinger
Resistance to social pressure
Evans
Goal setting/Feedback
Locke & Latham
Kiesler
Commitment/Cognitive dissonance
Zimbardo
Prompts
Anticipated regret
Richard
Attributional theories, Weiner
Attributional dimensions
Relapse prevention, Marlatt & Gordon
High-risk situations/Coping
Education Research, 8, 255-269.
Stake RE 1995 The art of Case Study research, Thousand Oaks: Sage Publica
tions.
Strauss & Corbitt, 1991
Strecher et al., 1 995]
Stroebe et al., 1993
Suls & Wils, 1991
Turner, 1991
'I
Tversky & Kahneman, 1986].
Van der Pligt, J., Otten, W., Richard, R., & Van der Velde, F. (1993). Perceived
risk of AIDS: Unrealistic optimism and self-protective action. In B.J. Pryor, &
G.D. Reeder (Eds.), The social psychology of HIV infection (pp. 39-58). Hilsdale,
NJ: Erlbaum.
Van Raaij et al., 1988
I
Additional theories on implementation
Diffusion theory, E.M. Rogers
Characteristics of the innovation
Diffusion theory, E.M. Rogers
Dissemination/Adoption/
Implementation/Continuation
Orlandi
Linkage
Weiner, B. (1986). An attributional theory of motivation and emotion. New
York: Springer.
Additional theories on environmental change
community development
social action
coalition formation
empowerment
P. Veen 1 985
Weinstein, N.D. (1988). The precaution-adoption process. Health Psychology,
I
7, 355-386.
Wilde, 19861
J
Windsor et al.
Winn, 1991
Woolcot
55
Position: 1797 (2 views)