106 Applying Social Psychology
The Help Phase:
Once the factors causing the outcome variable have been identified and mapped in the process model, the intervention can be developed. An intervention is a means to change the causal factors and thus the outcome variables in the desired direction. An adequate intervention targets one or more causal factors in the process model. Yet often it is not feasible or even necessary to target all variables in this model. Therefore, the first step in the Help stage of the PATH model is to determine which causal factors will be targeted in the intervention. The modifiability of the factors and the expected effect sizes of interventions will direct this choice. Once these factors have been identified, an intervention that targets these factors can be developed. Decisions must be made about how the target group will be reached and what the content of the intervention will be. The content depends largely on empirical evidence. The last step in the Help phase concerns the implementation process. Here care is taken that the intervention is used as intended. We want to emphasize that the present chapter only gives an introduction into the art of intervention development, and that more detailed approaches are available elsewhere, for instance, in the context of health education (see for example, Bartholomew, Parcel, Kok & Gottlieb, 2006).
It is known that racial discrimination may have severe psychological consequences for people who are discriminated against: they may feel unfairly treated, depressed, angry and/or sad. But can racial discrimination also harm physical health? In general Black US citizens suffer from higher blood pressure than White US citizens. Is it possible that this has something to do with racial discrimination? To answer this question, researchers Nancy Krieger and Stephen Sidney conducted a study* among more than 4000 Black and White adults. Participants' blood pressure was assessed, and they were asked about their experiences with racial discrimination and their ways of coping with it. They were asked, for Box 5.1 Case Study: Racial Discrimination
and Blood Pressure
and Blood Pressure
The Help Phase 107
PREPARING INTERVENTION DEVELOPMENT
The essence of the final Help phase is that interventions must focus on changing factors in the explanatory model. It is not always necessary, appropriate, or possible to target all the factors in the explanatory model. Therefore, the applied social psychologist chooses the factors that are modifiable and that have the greatest effect on the outcome variable. To do so, it is convenient to put all the factors from the process model into a balance table (see Table 5.1).
Although presumably many variables in the selected process model can be influenced, there may be considerable differences in the degree to which this is possible. Three questions can help to exclude factors that are difficult to change:
108 Applying Social Psychology
1. Does the factor concern a stable personality trait? For example, when a psychologist wants to develop an intervention to tackle shyness, one may include introversion as a personality variable with a high degree of explanatory power in the model, but this variable has little potential for change outside intensive psychotherapy. Or when one includes neuroticism as a variable in a model predicting burnout, one needs to realize that this is a stable personality trait that will be difficult to change.
2. Is the factor related to deeply held political or religious values? For example, it will be virtually impossible to attract attention for a programme on condom use from people who, based on their religious convictions, are strongly against premarital sex. Or it may be difficult to convince selection officers with very negative attitudes towards immigrants to endorse a policy to employ people from minority groups.
3. Is the factor related to stable environmental conditions? For example, a problem might be that students do not park their bikes in the appropriate parking spaces at university. Perhaps they see many other bikes parked in inappropriate spaces and follow that example. However, this might be due to insufficient bike storage facilities on campus, which is a stable environmental condition.
Not all factors in the process model have an equally strong impact on the outcome variable, and applied psychologists should focus on the ones that have the strongest effect. This selection is facilitated if there is empirical evidence for the strength of the causal relationships in the model. Often, however, this is not available and psychologists 'guestimate' the effect sizes. Various sources of information can be helpful to estimate the effects.
Past Experience with Similar Situations
Suppose a school aims to tackle cultural segregation among their students. In the
process model, the recruited psychologist identifies 'knowledge about people from other cultures' and 'personal contact with people from other cultures' as factors predicting segregation. Yet last year the school board decided to provide youngsters with positive knowledge on other cultures, and, although students were more positive, this seemed to have no effect on social interactions in the school. Armed with this knowledge, it does not seem very sensible to try this strategy again.
For a number of factors in the model, there may be empirical evidence that they are resilient to change. For example, the literature shows that most people are unrealistically optimistic about their lives (Weinstein & Klein, 1996). Most children, for example, believe they have less risk than the average child of becoming overweight and developing health problems as a result of being overweight. As this optimism is a statistical impossibility — it simply cannot be that most people are better off than the average person — a psychologist may believe it could help to educate people about this illusion. However, research suggests that such biased perceptions are hard to correct and that their influence on behaviour is limited (Weinstein, 2003). Thus, one would not select this factor to be targeted in an intervention. In general, it makes sense to look for evidence showing the degree to which the model variables are changeable.
The Help Phase 109
Figure 5.1 Process model: What factors influence weight regulating parental behaviours?
The Balance Table
A balance table helps in making the decision on which factors will be targeted in the intervention. We illustrate the balance table through the problem of obesity among children in the UK. Britain is one of the 'fattest' nations in Europe. In 2000, 27 per cent of girls and 20 per cent of boys aged 2 to 19 years were overweight (see the websites on NationalStatistics, www.statistics.gov.uk). Obesity amongst children is a problem of great concern: such children have a high risk of developing long-term chronic conditions, including adult-onset diabetes, coronary heart disease, orthopaedic disorders, and respiratory disease. A social psychologist is asked to develop an intervention to help tackle this problem. In studying the literature she discovers that one of the major causes of obesity is the influence of parents on the weight of their children (Jackson, Mannix & Faga, 2005), namely, that parents may not do enough to stop weight gain in their children. Such behaviours are referred to as weight regulating parental behaviours. A psychologist will develop a process model in which these behaviours are the outcome variable (see Figure 5.1).
Next, a psychologist will evaluate all the variables from the process model with regard to their modifiability and their effect size, that is the magnitude of the impact of the change on the outcome variable.
110 Applying Social Psychology
Table 5.1 Balance table
Variables from the process model
1. Perceived negative outcomes of child being overweight
2. Perceived positive outcomes of weight reducing parental behaviours
3. Perceived response-efficacy of child's behaviour to lose weight (does a change in the child's behaviour lead to losing weight?)
4. Perceived control over weight reducing parental behaviours
5. Knowledge on weight reducing parental behaviours
6. Perceived control over child's behaviour
7. Experience with structured parenting
8. Positive attitude with regard to democratic parental style
0 or +
Note With regard to modifiability: ++ = high modifiable; + = medium modifiable; 0 = low modifiable, — = not modifiable, +/0 = depends on another variable.
With regard to the effect size: ++ = large effect; + = moderate effect; 0 = small effect; — = no effect +10 = depends on another variable.
First, she evaluates the modifiability of the eight causal factors in the process model. The first three factors — the perceived negative outcomes of a child being overweight, the perceived positive outcomes of regulating one's child's weight, the perceived response-efficacy of a child's behaviour to lose weight — are beliefs based on factual knowledge and on interpretations of past events or experiences. In general, beliefs can be influenced quite well. Furthermore, the perceived positive outcomes of weight regulating parental behaviours can only be brought about under conditions of sufficient response-efficacy regarding the child's behaviour, that is, when the parental behaviours produce the desired change in the child's behaviour. The knowledge about weight regulating parental behaviour, and how to perform it, can also be modified as it only requires adequate basic information processing and storage. Structured parenting refers to a parenting style in which children are actively guided and given clear directions for choices and behaviours, for example, concerning food intake and physical exercise. The experience with this type of parenting is also modifiable and can be changed by practising it. Structured parenting is less likely the more parents tend to engage in a democratic parenting style, i.e. have a parenting style in which children are stimulated (or often just left) to make their own choices. A positive attitude with regard to a democratic parental style is perhaps difficult to change as it may be based on parental modelling and a history of perceived reinforcement of that style. The conclusions with regard to the factors' modifiability are depicted in Table 5.1.
Next, she will evaluate the effect size of the causal factors in the process model. How strong is the effect on the outcome variable? Changes in the first three variables in the balance table are probably possible but their effects on the outcome variable largely depend on parents' perceived control over regulating their children's weight. Just changing these variables may have a small effect as only some parents will have
The Help Phase 111
sufficient control beliefs. Changing the knowledge on weight regulating parental behaviour can have a large effect, as it is a primary condition to engage in such behaviour. Changing the positive attitude with regard to a democratic parental style has uncertain effects, as it does not guarantee that an adequate alternative style will be adopted. Changing the perceived control over the child's behaviour can have large effects as it is a basis for developing perceived positive outcomes of regulating the weight of one's children and perceived control over this behaviour. Finally, experience with structured parenting can have various effects but it does not guarantee that parents are skilled in the specific behaviours regulating the weight of their children. A psychologist will summarize her findings in the balance table (see Table 5.1).
From the balance table it appears that an intervention that targets the knowledge on weight regulating parental behaviour and the perceived control over a child's behaviour will probably be most successful. In addition, a psychologist may also target the perceived negative outcomes of a child being overweight, the perceived positive outcomes of the parental behaviour, or the perceived response-efficacy of a child's behaviours.
DEVELOPING THE INTERVENTION
Once the psychologist has decided which variables to target, the intervention can be developed. Three tasks can be distinguished in the development of an intervention:
1. Choosing the right channel, in which way one may reach the target group members for example.
2. Selecting the appropriate methods, the way the changes will be brought about, for example, by offering a role model or performing a skill exercise.
Developing the strategies, the translation of the methods into concrete aspects of the intervention. For example, when the method is social modelling, the strategy refers to the exact model and the things the model says and does.
Developing the strategies, the translation of the methods into concrete aspects of the intervention. For example, when the method is social modelling, the strategy refers to the exact model and the things the model says and does.
The channel, the method and the strategy must consider the target group for intervention. A social psychologist may focus on improving patient skill in taking a specific medicine when the target group consists of those patients who already use, or who will use, that particular medicine. The choice of channel is guided by the need to reach this target group (for example, through using a pharmacy) and by the method (for example, modelling: watching another patient taking the medicine on a DVD). The specific model (strategy) demonstrating the skills depends upon the target group (for example, using an older model when the patients are elderly). It is important to note that the development of an intervention is usually a dynamic process: choices for the channel, the method and the strategy are made in combination (see Figure 5.2).
112 Applying Social Psychology
Figure 5.2 The development of an intervention: channel, method and strategy
given of various channels and their characteristics. Channels have several features, and may vary from simple (for example, sticker or label) to complex (for example, community intervention), each communicating a distinct type of information (for example, text, picture) and a different volume of information (for example, one simple message versus a complex set of arguments). Some channels communicate with high intensity (say, group therapy) and others with low intensity (say, information signs).
Channels also differ in the potential reach of the target group (see Table 5.2). For example, a label on a product has the potential to reach all the users of a product, while a radio message reaches only part of the population of users. In addition, some channels will only have small effects on the individual level (say, a sticker), while others can have large effects (the example of group therapy). Lastly, channels may bring about different types of effects. For example, a label is appropriate to increase people's knowledge, while counselling is more appropriate to change complex and deeply rooted behaviours.
The channel is chosen on the basis of information about the target group, the relevant variables, methods and strategies. The following issues should be considered when choosing the channel:
1. Is the channel an effective way to reach the target group? (See Table 5.3, potential reach.) When people want to know how to use a product, say a wrist-exercise tool, using the label on the tool works better than a television ad. The label ensures that people who buy the tool have access to the information.
2. Is exposure through this channel intensive enough to change the variable? (See Table 5.2, effect on individual level.) A billboard depicting a young woman during a physical work-out may remind people that physical exercise is desired, but may not lead to a change in attitudes toward working-out. Daily feedback through the internet, however, may shape people's positive experiences with fitness, leading to the desired changes.
Table 5.2 (Continued)
application2 Psychological change/
· Internet/E-mail Information sites,
interactive sites, People with access to Small New knowledge/
reminder mails, etc the internet Psychological/
· Cell-phone Reminder calls, SMS People in the database Small Psychological
or all those who request change/
the service Behavioural
· Television Commercials, infomercials, People who watch TV Medium New knowledge/
documentaries, spots, etc channel at the time Psychological
of broadcasting change
· Minimal A few short personal contacts People who are Medium Psychological
counselling referred to it change/
· Extensive Several or many personal contacts of People who are Medium/Large Psychological
counselling about 30-60 minutes referred to it change/Behavioural
· Group training Education and skills training in a group; People who are Medium/Large Psychological
mostly 1 to 10 meetings referred to it change/Behavioural
· Group therapy Applying therapeutic means to induce People who are Large Psychological
change in individuals partly by means of referred to it change/Behavioural
the group; often many meetings change
Table 5.2 (Continued)
· Structural environmental changes
· Community' intervention
Agreements about permitted
or banned behaviours in more or less specified contexts
Changing the environment to regulate experiences or behaviours, e.g. by regulating the exposure to certain stimuli or the availability of a specific product Changing peoples' experiences or behaviours by changing the structural and informational influences of the community they belong to
Depends on the application'
encounter the specific environment
Members of the community
New knowledge/ Psychological change/ Behavioural
Stickers can be used in many different ways. For example, they can be distributed freely and then much depends on those people who encounter the distribution point. Stickers can also be distributed together with product X, exposing only those people who buy that particular product.
A CD-Rom/DVD can be distributed in many ways. For example, they can be actively sent to people who are registered in a database, they may be ordered by those who feel they have a need for it or they can be distributed together with product X.
There are many types of regulations and laws and they may be applied broadly or may only be relevant in specific situations. ° Community interventions are a composition of various channels developed to influence the same factors.
116 Applying Social Psychology
3. Is the channel appropriate for the method and strategy? A sticker is less appropriate for modelling complex skills, such as learning to lead a healthier lifestyle, while an interactive DVD gives several possibilities for modelling and practising healthy skills. (See Table 5.2, effect type.)
Intervention methods also require consideration. Methods are often derived from theoretical frameworks. For example, the foot-in-the-door technique (Cialdini & Trost, 1998), according to which people more easily accept a major request after first complying to a minor request, is embedded in the theory of self-perception that argues that people adjust their attitudes to their behaviours (Bern, 1972). Such theories are important because they specify the conditions under which the method is most or least likely to be successful. For example, according to social learning theory, modelling is most effective when the similarity between model and target individual is high (Bandura, 1986). From the various theories, phenomena and concepts in the Glossary (pp. 136-47) one can often deduce ideas about methods.
Selection of a method depends, first, on consideration of the balance table (see Table 5.1). For each variable, an intervention method must be chosen. Suppose 'attitudes of police officers towards foreigners' and 'communication with foreigners' are the selected variables to improve the treatment of tourists in a city in Spain. Modelling might be used to demonstrate communication skills, whereas the method of argumentation might target attitude change. Second, selection of a method depends on the extent to which the method 'fits' the variable one aims to change. Whereas some channels can motivate people to show the desired behaviour, they cannot teach them how to change it. For example, it is easy to arouse fear in smokers through a 30 second television advertisement, but it is difficult to help them quit smoking using this channel. In contrast, it is
The Help Phase 117
often sufficient to remind people of the benefits of a certain behaviour by using a prompt. For example, a sign on an elevator door can prompt people to use the stairs for exercise.
The following methods are frequently used in psychological interventions.
Setting concrete and specific goals is important. Goals direct people's attention and effort, provide them with expectations, and give the opportunity for feedback on goal accomplishment, thereby regulating motivation. Goal setting changes behaviour by defining goals that people must reach in a given period of time (Locke & Latham, 2002). For instance, as regards being overweight, a goal can be set in terms of weight loss in kilos over a particular time period. Sub-goals can help people work on small, but important, steps in reaching the superordinate goal. For example, in patients who have had a stroke, a goal in the rehabilitation could be: 'After three months I can walk 200 metres by myself'. Many studies support the effectiveness of goal setting. Evans and Hardy (2002) examined the effects of a five-week goal-setting intervention for the rehabilitation of injured athletes. The results showed that a goal-setting intervention fostered adherence and self-efficacy. McCalley and Midden (2002) provided participants with feedback to increase household energy conservation behaviour. They showed that participants who had set goals for themselves eventually saved more energy.
Fear communication can be effective to motivate certain behaviours. For example, health behaviours, such as using condoms, can be encouraged by graphic information about sexually transmitted diseases (Sutton & Eiser, 1984). An interesting study by Smith and Stutts (2003) compared the effects of a fear appeal showing the cosmetic effects of smoking (unhealthy looking faces) with the effects on one's health (cancer). They found that both fear-appeal conditions effectively reduced smoking. It must be noted that fear communication is only effective (and ethically justified) when it is accompanied by explicit guidelines on how to avert the health threat.
Modelling refers to learning through the observation of others. Watching others behave and showing the consequences can teach people to perform a new behaviour (Bandura, 1986). Modelling is useful for all kinds of skills, for example, coping with criticism, cooking meals, and using condoms. In a meta-analysis, Taylor, Russ-Eft and Chan (2005) examined the effects of different types of modelling. Do people learn more when a skill is modeled positively (showing what one should do), negatively (showing what one should not do) or in combination? These psychologists concluded that skill development was greatest when role models were mixed, that is, when they showed what one should as well as what one should not do.
118 Applying Social Psychology Enactive Learning
The most effective way of learning a skill is to try to accomplish it yourself. This is called enactive learning. In interventions, people can be stimulated to practise a certain skill and evaluate it. For instance, to foster students' interest in science subjects like mathematics, Luzzo and his colleagues (1999) exposed students to either a video presentation of two university graduates discussing how their confidence in maths had increased (so-called vicarious learning) or a maths task providing these students with feedback on their maths skills (enactive learning). The enactive learning condition proved more effective.
Social comparison — information on how others are doing — may affect one's mood and well-being (Buunk & Gibbons, 2007; Festinger, 1954). For example, in the context of coping with cancer, Bennenbroek et al. (2003) provided cancer patients undergoing chemotherapy with social comparison information to increase the quality of their life. The intervention consisted of a tape recording of fellow patients telling their personal stories about either the treatment procedure, emotions experienced during the treatment, or a tape about the way they tried to cope with the situation. The latter tape especially reduced anxiety over the treatment and improved patients' quality of life.
Implementation intentions are intentions to perform a particular action in a specified situation (Sheeran, Webb & Gollwitzer, 2005). Sometimes people are asked to formulate their implementation intentions. For a person who wants a low fat diet an implementation intention could be: 'When I am at the supermarket I will put the low-fat butter in the shopping trolley', or 'When I am at the party on Friday night and somebody offers me cake, I will decline'. Asking people about their implementation intentions may increase the occurrence of desirable behaviours. Steadman and Quine (2004), for instance, showed that asking participants to write down two lines about performing testicular self-examination led to the desired action. Likewise, Sheeran and Silverman (2003) compared three interventions to promote workplace health and safety and found that asking people to write down their implementation intentions was the most effective.
Reward and Punishment
In general, people repeat behaviours that are followed by a positive experience (reward) while avoiding a negative experience (punishment). In the smoking example, a reward may take the form of a refund for the costs of a smoking cessation course from the health insurance company, if people quit smoking for at least three months. In contrast, the government may punish people for smoking by increasing the price of cigarettes. Furthermore, people can learn to reward or punish themselves. For example, people who succeed in refraining from smoking for a week could treat themselves to a cinema visit. In general, there is much evidence for the effects of punishment and rewards. Punishment of undesirable behaviours (for example, high fines for drunk driving) works
The Help Phase 119
best when it is accompanied by rewards for desirable behaviours (for example, praise for staying sober before driving) (Martin & Pear, 2003).
Feedback on accomplishments is essential in behavioural change. People losing weight want to know how much weight they have lost. Without feedback people become uncertain and their motivation deteriorates because they do not know whether they have made progress (Kluger & DeNisi, 1998). Brug and his colleagues (1998) provided people with tailored computer feedback on their diet (vegetables, fruit and fat intake) and on dietary changes. Both feedback types appeared to improve dietary habits. Likewise, Dijkstra (2005) showed that a so-called fear appeal to smokers — a single sentence of individual feedback (It appears you are not aware of the changing societal norms with regard to smoking') — was twice as effective in reducing smoking as no feedback.
Box 5.2 Interview with Professor Gerjo Kok of the University of
Maastricht (The Netherlands)
Maastricht (The Netherlands)
One of the oldest and most prominent application areas of social psychological theory is health. Professor Gerjo Kok is one of the leading scientists in this field.
'In addition to research on the causes of unhealthy behaviour, we have developed a process protocol, called Intervention Mapping, that provides guidelines and tools for the development of health promotion programmes. In certain ways
120 Applying Social Psychology
the protocol is like this book. It helps social psychologists, health organizations and/or the government to translate social psychological theory and research in actual health programme materials and activities and develop health intervention programmes that are maximally effective. In general I see a great future for applied social psychology. I believe that social psychology will play a growing role in the solution of societal and health problems. Especially with regard to the study of unconscious processes (such as habits), the field of group dynamics, and the study of environmental determinants of behaviour (such as social norms and social control) I expect social psychologists to become (even) more active.'
Interested in Gerjo Kok's research? Then read, for instance:
Dijkstra, A., De Vries, H., Kok, G. & Roijacker, J. (1999). Self-evaluation and motivation to change: Social cognitive constructs in smoking cessation. Psychology& Health, 14(4), 747-759.
Kok, G., Schaalma, H.P., Ruiter, R.A.C., Brug, J. & van Empelen, P. (2004). Intervention mapping: A protocol for applying health psychology theory to prevention programmes. Journal of Health Psychology, 9, 85-98.
Methods have to be translated into a specific strategy. The strategy is the actual intervention people will get exposed to. For example, using television as the channel and modelling as the method, the strategy would specify the age and gender of the role model. In the case of flu vaccinations for elderly people, the strategy might be a television spot ad in which viewers watch an older woman, with good health, being interviewed in her doctor's waiting room, before having her vaccination.
To come up with strategies, a global intervention plan could be made specifying the methods, channels, target groups, and variables to be changed as identified on the basis of the balance table. Here are some examples:
· Modelling (method) on television (channel) to motivate women with overweight children (target group) to monitor their children's body weight (variable to be changed).
· Giving feedback (method) through the internet (channel) about the length of time youngsters (target group) engaged in exercise during the past week to support an increase in their physical stamina (variable to be changed).
· Offering arguments (method) to motivate quitting smoking (variable to be changed) in a self-help book (channel) for smokers of all ages (target group).
· Repetition (method) of the word 'action' in the text of a model (method) presented in a leaflet (channel), designed to motivate obese people (target group) to formulate implementation intentions with regard to reserving a seat with extra space (variable to be changed) on international flights.
The Help Phase 121
Next, based on these global intervention descriptions, the social psychologist could select a strategy for intervention. This usually takes place in two phases, a divergent and convergent phase. In the divergent phase, the psychologist lists as many strategies as possible. In the convergent phase, he or she critically evaluates these strategies.
The Divergent Phase: Inventing Strategies There are various techniques to generate interventions:
· Direct intervention association. Ideas for strategies can be based on all kinds of sources, such as what one has seen on television, what makes intuitive sense, what has proven to be effective in the literature or what people have experienced themselves.
· Direct method approach. This approach consists of looking at strategies that have been used in similar situations. For example, suppose that the global intervention description is: 'Provide information on the appropriate use of a new type of toothbrush on a label to prevent mouth injury in patients with bad teeth.' A psychologist might then inspect existing labels on toothbrushes. Also labels with regard to other devices that could injure people could be used to generate ideas.
· Debilitating strategies. This approach is to come up with strategies that have undesired effects on the problem. In the case of the global intervention description: 'Modelling on television to motivate women to monitor their children's weight', the model should probably not be a retired millionaire on his ranch. By generating ideas of what would probably have no or reverse effects, we can learn about what would have an effect, about the relevant dimensions of an effective intervention and about ways to operationalize the strategies.
· Interviews. Interviewing people from the target group could generate additional ideas for strategies. With the global intervention description, 'Modelling on television to motivate women to monitor their children's body weight', a woman with young children might be interviewed about her preference for role models that might inspire her. Here is an example of such an interview:
'If there was to be an intervention on television in which a person tried to convince you that monitoring your children's body weight is important, what kind of person would you trust most?'
'I think I would be persuaded most by someone with experience of the problem. It should be a mother but knowing how things are manipulated on television I would need to have proof that she really is a mother with experience.'
'Do you have any other ideas about the person and what the person would say that would help you to accept the message?'
'The mother should be a sensible person, with some education. I think she should also be serious about the topic; after all, it is about the health of your children.'
'What kind of person would you trust least?'
'When I got, one way or another, the impression that they are indirectly
trying to sell a commercial product I would immediately stop watching.'
trying to sell a commercial product I would immediately stop watching.'
This kind of interview — asking for the desired but also the undesired characteristics —can generate new perspectives and ideas about strategies. From the above, we learn that people might feel they are being manipulated, which should of course be avoided.
122 Applying Social Psychology
· Insights from theory. This approach consists of looking at relevant social psychological theories. With regard to the method of goal setting, for instance, the difficulty of the goal is crucial (Strecher, Seijts, Kok, Latham, Glasgow, DeVellis, Meertens & Bulger, 1995). In general, goals stimulate performance when they are difficult and offer a challenge but at the same time are within someone's reach. Losing two pounds in two months might not motivate a person much, because the outcome is not very attractive, but losing 20 pounds in two months might be unrealistic. Thus, in developing strategies, the psychologist should look carefully at what the theory predicts.
· Insights from research. This approach consists of looking at relevant social psychological research. For example, research shows that people are less defensive with regard to processing threatening information (for example on cancer risks) when their self-esteem is boosted (Sherman, Nelson & Steele, 2000). Therefore, in developing a fear-appeal we might want to include a self-esteem boosting method, for example, asking people to write an essay on the good things they have done recently (Reed & Aspinwall, 1998). Such 'manipulations' are described in the method sections of empirical articles and can provide the social psychologist with creative ideas for strategies.
The Convergent phase: choosing the strategy
The divergent phase often results in a laundry list of strategies. Therefore, a limited number of strategies need to be selected. The choice for a particular strategy or set of strategies must have both a theoretical and empirical basis. First the strategy should take into account the conditions underlying the theory. For example, in the
case of modelling, the theory specifies that the actual model must be similar or at least relevant to people in the target group (Bandura, 1986). Second, it is preferable that the choice of strategy is based on empirical evidence from either laboratory experiments or field studies. Ideally, evidence ought to be available for the combination of the channel, the method, the strategy, the variable to be changed and the
target group. For example, for the global intervention description: 'Modelling on television to motivate women to monitor their children's body weight', the strongest evidence would come from a field experiment in which such an intervention was tested in a specified target group against a control condition. Somewhat weaker evidence would come from testing the intervention video in the laboratory. The stronger the empirical evidence for the intervention, the higher the chances that the intervention will indeed be effective.
Sometimes evidence for the effectiveness of a certain strategy is simply not there. In that case, especially when the costs of an intervention programme are high, we recommend that the effectiveness of a new strategy should first be tested through research.
BUILDING THE INTERVENTION PROGRAMME
The Help Phase 123
graphic designer who is acquainted with the technological possibilities, such as paper sizes, colour use, lay-out, visual angles, and dynamic effects. Here are some rules of thumb for preparing materials, based on our own experiences:
· Be as specific as possible. In the case of a leaflet, formulate the final arguments, write the introduction, link the arguments, and choose the font size and type. In the case of a video with the method of modelling, write the script and include what should be said and done and what should happen in the video.
· In the case of an intervention with several channels (for example, billboards and television spots) or sequential elements (for example, group counselling sessions), all parts must be fine-tuned and a protocol must be written as well as with planning the intervention.
· If professional artists are involved, it should be clear how much influence they can have over the end-product. The communication with professional artists should be highly interactive and several versions may have to be designed by the artist in order to come up with a product.
· The intervention often includes more than one strategy. In principle, all aspects of the intervention that can be read, seen or heard should be part of a strategy. Thus, the colours, the sizes, the sounds, the timing, the wording, the movement, the background, the aspects of the background, and the specific shapes should all refer to an identifiable strategy. One way to test this is to point to a single aspect of the intervention and ask: What strategy is this part of and what is the method of operationalization?'
Pre-Testing the Intervention
Each planned intervention must be pre-tested. The primary function is to improve the intervention and to avoid major flaws in the design. A pre-test does not necessarily include a behavioural measure. It primarily ensures that the target group will attend to the message as well as understand the message. For example, to assess if people from the target group attend to the persuasive message, they may be asked: 'Did you find the information interesting?', 'Why did you find it not interesting?', and 'Were you still able to concentrate on the message at the end?' In addition to such general questions, one may add more specific questions. For example, when the social psychologist has chosen a role model who is trying to persuade members of the target group, there may be questions like 'How similar do you feel to the person in the video?', 'How sympathetic do you find the person in the video?', 'Did you believe the person on the video indeed suffers from disease X, which he claims to do?', and 'What aspects of the model should be changed for you to believe the person?'. The format of the pre-test usually includes exposing a few target group members to the preliminary intervention and assessing their reactions. This assessment can be done in different ways.
· Interview. This is in general a useful method to pre-test the intervention. One may have interviews with individuals from the target group and ask questions like the ones above. In addition, one may ask more specific questions. For instance, people may be asked to read a leaflet and tell the interviewer about their reactions, how reliable they found the information, how realistic they found it, and what they liked or did not like about the content or layout.
124 Applying Social Psychology
· Quantitative assessment With this type of pre-test, people from the target group answer closed questions about the intervention in a questionnaire. For example, they may be asked to rate the reliability of the source on a 7-point scale (from 'not at all reliable' (1) to 'very reliable' (7)) or they may be asked whether the intervention 'took too much time' , 'was just right' or 'was too short'. With a more experimental paradigm, when one wants for example to use a sticker to indicate the location of the fire-extinguisher in the building, this sticker could be pretested by comparing different versions.
· Recall. The psychologist may also use a recall task, which assesses which aspects of the intervention people from the target group remember after having being exposed to the intervention. This might give an insight into which strategies have the highest salience. Imagine a billboard depicting a celebrity promoting safe sex, but when individuals from the target group are asked to recall the characteristics of the billboard, half of them only remember the name of the celebrity and not what he or she was promoting. In this case, the salience of the messenger has apparently distorted the message of the intervention.
· Observation: People from the target group may also be observed while being exposed to the intervention. For example, in the case of testing a billboard, eye movements may be monitored to track which aspects of the billboard they pay most attention. Likewise, when testing an internet website, the link-choices and the time spent on each page might be monitored.
· Expert opinions For pre-testing the intervention, one may also ask the experts involved in bringing about the effects of the intervention. (For example, in the case of a leaflet to increase treatment adherence a doctor might be asked to give his opinion.) Or a shop-keeper who is supposed to hand out a leaflet to everyone who buys product X may be asked whether he or she thinks people will indeed look at the leaflet.
In general, participants seldom agree completely about an intervention in a pre-test. Therefore, an applied social psychologist must not only look at the pre-test data, but must also consider theoretical aspects as well as empirical evidence that may be relevant. After revisions have been made, the improved intervention can be pretested a second time. The final version of the intervention can now be developed and distributed.
IMPLEMENTATION OF THE INTERVENTION
When the intervention has been developed the implementation process can start. The implementation process has one major goal: to ascertain that the intervention is used as intended. Thus, when a psychologist develops an intervention campaign with leaflets and television advertisements, members of the target group must be exposed to these messages. When all members are exposed to the intervention (for example, they have all read the leaflet or have at least seen one television advertisement), the intervention is implemented optimally. Note that implementation is not about the effects of the intervention but about positioning the intervention in such a way that it can have its effects.
The core challenge of the implementation phase is that the extent to which the target group is exposed to the intervention depends on the people and organizations that are involved in the distribution of the intervention. For example, with regard to a leaflet
The Help Phase 125
Figure 5.3 Proportion of people who report needing the intervention programme, who are aware of its existence, who have started using it and who have completed the programme
about medicine intake, pharmacists may be involved by motivating their employees to distribute the leaflet to all patients getting a specific medicine. We cannot expect that all these people are as motivated to get the target group members exposed to the intervention as the initiators and developers of the intervention. Therefore, the implementation of an intervention involves motivating others and removing any perceived obstacles to allow them to engage in their specific tasks.
Sometimes people who help with the implementation are simply not aware of the intervention. Paulussen and his colleagues (1995; see also Paulussen, Kok & Schaalma, 1994) studied the implementation of an educational programme consisting of several lessons designed to promote AIDS education in classrooms. Almost all the teachers had initially expressed an interest in participating. Yet only 67 per cent of the teachers were aware of its existence on the curriculum and only 52 per cent initially started to teach it. Although Paulussen et al. (1995) did not assess whether the teachers finished a whole curriculum, this percentage is likely to be substantially lower (see Figure 5.3).
Thus, although psychologists may develop an excellent intervention programme, if only a few people are actually exposed to the intervention because professionals who are essential for its implementation, such as teachers or doctors, are not aware of it or do not use it properly, the impact of the intervention on the problem may be small or non-existent.
The Implementation Process
126 Applying Social Psychology
Figure 5.4 The diffusion process of innovations
large-scale changes in the use of an innovation (for example, using a new toothbrush or using a new method to quit smoking) take place in time. This process is referred to as the diffusion process. Rogers distinguishes four phases. We will illustrate the process here with an example of a leaflet for battered women to get professional help. This leaflet is to be handed out by general practitioners if they suspect domestic violence.
1. Dissemination phase. In this phase the general practitioner becomes aware of the existence of the leaflet on domestic violence and discusses it with colleagues.
2. Adoption phase. In this phase the practitioner becomes motivated to use the innovation and to hand out the leaflet to patients who are suspected to be victims of domestic violence.
3. Implementation phase. In this phase the doctor actually engages in the behaviour that will expose the target group to the intervention: he hands out leaflets to the right patients.
4. Continuation phase. In this phase handing out the leaflet becomes normal practice.
In stimulating the diffusion process, all four phases will have to be addressed: raising awareness among general practitioners, motivating such practitioners to detect target group members (i.e., women who may be victims of domestic violence), and to hand out the leaflet to these women, supporting the practitioners in the actual execution of the behaviours, and providing feedback and reinforcement to maintain the behaviours (for example, by calling practitioners on the phone and giving them support and advice).
Note that the diffusion process highlights the phases in the implementation process. It does not define all the parties that are involved in the implementation, except for the end users of the innovation (in the above example the general practitioners). The next step is to identify all the people and organizations involved in the implementation process. For each of the four diffusion phases, different people and organizations may be involved.
Mapping the Implementation Route
The Help Phase 127
involved in the implementation and their motivations and barriers to performing their task in the implementation are mapped. Developing an implementation route consists of three steps:
1. Mapping the actors. The route shows the actors in the relevant networks in which they communicate and their means of communication. For example, with regard to the implementation of: 'A leaflet with information and arguments for battered women to get professional help', individual general practitioners are members of a regional organization, which is part of the national organization for general practitioners. Within and between these levels of organization, individuals and organizations communicate via different means, including for example, formal meetings, professional journals and email. Furthermore, patients who read the leaflet and decide to seek professional help should be able to make a first appointment very quickly. Thus, the organizations providing professional help to battered women must also be involved in the implementation.
2. Assessing the motivations and the barriers for actors. For a successful implementation, each of the actors will have to engage in a specific task. For example, general practitioners should be motivated to engage in detecting battered women and should dedicate some time to this task. Furthermore, the board members of a general practitioners' organization should be motivated to invest some money in persuading general practitioners to perform the detection or to persuade insurers that the detection of battered women should be reimbursed. Thus the implementation route contains for all actors a diagnosis of the potential problems to engaging in the implementation's task and the specific barriers to performing it.
3. Identifying relevant policies. Besides identifying the actors involved, the relevant policies also need to be known. For example, it is possible that there is a policy for general practitioners that says that the general practitioner will not engage in detection tasks with regard to family matters. Although some general practitioners might still be motivated, this would rot be an ideal situation for the implementation of the intervention. Or it may be that there is a law that says that the police can only offer protection to a battered woman when there is objective evidence of domestic violence. This might inhibit women from seeking professional help. This law would counter the desired effects of the intervention structurally.
The Implementation Plan
When the actors, organizations and policies have been identified and the motivation and the perceived barriers to the actors have been mapped, the implementation plan can be developed. The implementation plan consists of all the steps that should be taken to stimulate the actors to conduct their task(s) in the implementation. In developing an implementation plan, the social psychologist must take two factors into account.
For each actor or level of actors and for each of the four diffusion phases, goals may he
formulated. For example, a goal for general practitioners in the first diffusion phase
could be: '80 per cent of the practitioners heard about the existence of the intervention
formulated. For example, a goal for general practitioners in the first diffusion phase
could be: '80 per cent of the practitioners heard about the existence of the intervention
128 Applying Social Psychology
material on domestic violence and at least 50 per cent discussed the material with colleagues'. On the organizational level, organizational goals should be formulated. For example, a goal in the adoption phase could be: 'The board of the national organization for general practitioners has decided to set aside one article in the professional journal on domestic violence and to develop a pre-publication on it in their communication with the regional organizations'. In principle, the goal should be that every actor has a positive attitude towards the implementation, or perceives their task in the implementation as a legitimate part of their job.
Next, the implementation plan specifies all the actions that must be taken to reach the goals. This implementation manual specifies how the goals can be reached. For example, the above goal with regard to the awareness of general practitioners of the intervention materials may be reached by actions directed at their national organization. For example, we might want the board to be motivated enough to decide that some articles on the detection of battered women should be published in their professional journal.
The Actual Implementation
The actual implementation exists through executing the implementation plan. Thus, all kinds of actions will have to be taken to inform and motivate the actors and to take away perceived or actual barriers for actors and to support the implementation. Actors may receive information designed to motivate them, or permission to act from a higher level in their organization or the means, in time or money, to do their part in the implementation. To support the execution of the implementation tasks, the actors might be contacted by e-mail, by letter, by telephone, by advertisements in professional journals, by presentations at meetings or by their internal communication channels.
As may be clear by now, the actual implementation of the intervention is time-consuming and much work has to be done before any target group members will be exposed to it.
To assess whether the problem that was targeted has indeed changed for the good, the last step in this intervention-development cycle is to evaluate the effects of the intervention. At least three types of evaluation are important: the effect evaluation, the process evaluation and the cost-effectiveness evaluation.
In the effect evaluation, the extent to which variables that are directly related to the problem have changed over time is assessed. At the very least the effect of the intervention
The Help Phase 129
on the specified outcome variable in the process-model should be assessed. There are, however, more outcome variables that may be evaluated to determine the effectiveness of the intervention. Imagine the case of the problem of obesity in which the level of exercise is the variable that the social psychologist aims to influence. The primary outcome variable in the process model is the level of exercise. However, the number of people who engage in sufficient exercise can also be a meaningful outcome variable. In addition, the percentage of obese people six or 12 months after exposure to the intervention could be an important outcome measure.
To assess to what extent the effects are temporary or permanent, an appropriate follow-up period must be specified. Long-term behavioural effects can be assessed 12 months after the exposure, although the 12 month period is based on consensus rather than on rationale. The best follow-up periods are based on specific arguments for the behaviour that is targeted. For example, because in smoking cessation most smokers who relapse do so within the first six months after the initiation, a six month follow-up should be sufficient; after this period very few ex-smokers relapse.
In the process evaluation, the elements that are preconditions for the intervention to be successful are assessed. There are two types of process evaluation. The primary-process evaluation refers to an assessment of the changes in the variables that underlie the changes in the outcome variables. For example, when the process model states that prejudice towards Muslims is caused by media misrepresentations of Muslims, the changes in prejudice as a result of unbiased publicity might be assessed in a primary-process evaluation. In principle, all the variables in the balance table that were targeted by the intervention(s) are primary-process variables and may be evaluated. The secondary-process evaluation refers to an assessment of the extent to which effective elements of the intervention have indeed been executed. For example, for individual counselling it may be essential that the counselor and the client develop a 'therapeutic relationship' because the therapeutic relationship serves as one of the methods of intervention. In a secondary outcome assessment, the extent to which the therapeutic relationship has been formed is assessed.
In a cost-effectiveness evaluation the costs of interventions are assessed and compared with the benefits. For example, obesity has huge societal costs specifically in terms of healthcare provision. If an intervention leads to a yearly decline of 500 people suffering from obesity, the healthcare savings can be calculated. A second aspect of the cost-effectiveness concerns the costs of intervention. The intervention development and implementation are costly as they involve professional labour and material costs. For a television advertisement to be broadcast, the costs for broadcasting must be paid. In the cost-effectiveness evaluation the savings caused by the intervention are compared to the costs of the intervention.
It is important that for the effect evaluation as well as the cost-effectiveness evaluation there are usually data sources available. Many commercial research agencies gather information on societal phenomena. such as the percentage of obese people and the number of unemployed. Thus, it may not always be necessary to gather additional data. On the other hand, it is important that outcome variables are carefully
130 Applying Social Psychology
Figure 5.5 Number of smokers (x 1000) 'as a function of type of channel' that were reached by the Millennium campaign 'I can do that too'
assessed. Therefore, a self-developed outcome assessment may be necessary. Especially with regard to the process evaluation, reliable measures must often be developed. (For further reading on the evaluation of intervention we would refer to other sources such as Action evaluation of health programmes and changes by John Ovretveit, 2001.) In the last paragraphs of this chapter we present a case study of a large-scale intervention that was successfully implemented to help smokers quit in the Netherlands.
Case Study: The Millennium Campaign 'I Can Do That Too'
In the Netherlands, the Dutch Expertise Center of Tobacco Control developed the Millennium campaign 'I can do that too' to reduce the percentage of smokers. The campaign consisted of a series of interventions through several channels to stimulate smokers to quit and support their attempt. The campaign started in October 1999 and ended in February 2000.
The intervention programme is depicted in Figure 5.5. In addition, the population was exposed to free publicity about the campaign. In the written media, no less than 519 articles were published on the Millennium campaign and 79 radio and TV items gave information on it.
The effectiveness of the campaign was assessed using a so-called panel design with measurement control groups (see Box 5.3). That is, before the campaign started, in October 1999 (Time 1), an initial measurement among smokers was conducted. This
The Help Phase 131
group constituted the panel group. It is common that such measurements may influence measurements done later with the same group. If one finds a change in the panel group, this could be an artefact of the first measurement (for instance, because it made them aware of the risks of their smoking habits). Therefore, when the second measurement was applied to the panel group (thus at Time 2, in February 2000) there was also a control group of smokers with no Time 1 measurement. The same was done for the Time 3 measurement (January 2001).
1. seven-days' abstinence (not smoking for at least the past seven days);
2. having engaged in an attempt to quit;
3. having positive intentions to stop smoking.
It was found that, at Time 2, those smokers who had watched the TV programme or the TV talk show at Time 1 had made significantly more attempts to quit. The long-term follow-up (Time 3) showed that smokers who knew the Millennium campaign also had more positive intentions to stop smoking. The researchers concluded that the Millennium campaign led more smokers to quit smoking and, for those who had not yet made an actual attempt, it had made smoking cessation a higher priority.
132 Applying Social Psychology
Developing an intervention programme includes the following steps: Box 5.4 The Help Phase: Developing an Intervention Programme
1. Make a list of all the causal variables in the process model and determine for each of them how modifiable they are and how large their effect will (probably) be. Make a balance table to summarize the results. Choose those factors for your intervention that are modifiable and that have the greatest effect on the outcome variable.
2. For each of the selected variables, come up with a channel, a method and a strategy to influence this variable. Justify your choices for the channels and methods and report on the use of different skills to look for appropriate strategies (i.e., direct intervention association, direct method approach, explore debilitating strategies, conduct interviews, look at relevant theories and research).
3. Reduce the potential list of strategies. In selecting suitable strategies take notice of the conditions underlying the theory that the particular strategy is based on and look for research that supports the effectiveness of that particular strategy. Develop the strategies you have chosen into an holistic intervention programme.
4. Pre-test the intervention programme by means of interviews, quantitative assessments, recall tests or observations.
5. Develop an implementation route. Map the actors involved in the intervention's implementation, assess the actors' motivations and the barriers they perceive, and identify relevant policies.
6. Develop an implementation plan. What steps have to be taken to mobilize and motivate the actors involved in implementing the intervention?
7. Implement the intervention programme and evaluate its effectiveness.
SUGGESTED FURTHER READING
Bartholomew, L.K., Parcel, GS., Kok, G & Gottlieb, N.H. (2006). Planning health promotion
programmes: An intervention mapping approach. San Francisco, CA: Jossey-Bass.
Ovretveit, J. (2001). Action evaluation of health programmes and changes: A handbook for a
user-focused approach. Abingdon, UK: Radcliffe Publishing.
Rochlen, A.B., McKelley, R.A. & Pituch, K.A. (2006). A preliminary examination of the 'Real Men, Real Depression' campaign. Psychology of Men & Masculinity, 7(1), 1-13.
Smith, K.H. & Stutts, M.A. (2003). Effects of short-term cosmetic versus long-term health fear appeals in anti-smoking advertisements on the smoking behaviour of adolescents. Journal of Consumer Behaviour, 3(2), 157-177.
Van Assema, P., Steenbakkers, M., Stapel, H., Van Keulen, H., Rhonda, G & Brug, J. (2006). Evaluation of a Dutch public-private partnership to promote healthier diet. American Journal of Health Promotion, 20(5), 309-312.
The Help Phase 133
A company that produces computer software asks you, as a social psychologist, for advice. The company consists of 10 departments, each of 50 employees, with every department managed by an executive. These 10 executives are in turn subordinate to a team of five directors that leads the company. Although 50 per cent of the employees are female, of the directors and executives only one person — the executive that runs the household department — is female. The team of directors asks you how they can improve the upward mobility of women in the company's hierarchy so the company will have more female leaders in the future.
(a) Read the following two articles:
Ritter, B.A. & Yoder, J.D. (2004). Gender differences in leader emergence persist
even for dominant women: An updated confirmation of role congruity theory.
even for dominant women: An updated confirmation of role congruity theory.
Psychology of Women Quarterly, 28(3), 187-193.
Eagly, A.H. & Karau, S.J. (2002). Role congruity theory of prejudice toward female leaders. Psychological Review, 109(3), 573-598.
Select from these articles causal factors and develop a process model. Make sure that you limit the number of variables to about 10 and don't take more than four steps back in the model.
(b) Estimate for each causal factor in the process model its modifiability and effect size. Make a balance table and select the causal factors at which the intervention should be targeted.
(c) For each of the selected factors, come up with possible strategies to influence this factor. Use direct intervention association and the direct method approach, explore debilitating strategies, conduct interviews, and look at relevant theories and research.
(d) Reduce the potential list of strategies by examining their theoretical and empirical basis.
(e) Outline a global intervention plan, a number of ways to present the intervention and make a plan for the implementation of the intervention.
(f) Develop an evaluation procedure to determine the effectiveness of the intervention programme
Abraham P. Buunk dan Mark Van Vugt