LB, Attitudes & Health, WEEK 4 Flashcards
What are attitudes?
- Attitudes are relatively stable predispositions towards (or evaluations of) socially significant ideas, people, events, objects, institutions etc (Allport, 1935) > attitudes are what you think about an object or perceive it (your evaluation of it) > An attitude is how you think/feel about something
- Attitude is not a behaviour in itself but it impacts behaviour > e.g: I think gyms are too public (attitude) vs I don’t go to gyms because they are too public (behavioural measure)
- Attitudes are socially significant meaning attitude is towards something that matters to the person or culture and not arbitrary matters. E.g.: attitudes towards political figures
How do attitudes work?
- Attitudes considered to have 3 components: emotional, behavioural & cognitive (including values & beliefs) > value = emotional valence applied to a topic beliefs = having ideas of something w/o concrete evidence towards it
- Attitudes can be explicit (conscious) or implicit (unconscious) but both guide our behaviour and decision making
- There is some degree of generalisability (however, remember that although people may have similar attitudes, the origin, strength, accessibility & manifestation varies from person to person)
- Attitudes allow us to evaluate things to make decisions quickly (Fazio, 1989) so they could reduce the strain on our cognitive resources ( + possibly keep us safe from an evolutionary perspective if an attitude allows us to identify a threating object) However, this isn’t always a good thing > attitudes could be harmful (e.g. knee-jerk reactions, prejudice, confirmation biases such as only looking for information which is consistent with our attitudes).
How are attitudes formed?
• Initially learnt during early socialisation > early interactions w/ caregivers + siblings where we learn typical ideas and preferences > this develops through the lifespan though
How can attitudes change through the lifespan?
• Anderson (1971,1981) made the Information Integration Theory, arguing that as we are exposed to new info, we weigh this up to see if it is consistent with existing attitudes, if it is inconsistent then we will change our attitude as a result
• A problem with this is that people tend to have confirmation biases and listen to the info which is consistent with their attitudes whilst ignoring the info inconsistent due to the cognitive strain
• Anderson also doesn’t explain how the person would change their attitude based on new info (do they just take the new info and agree with it or does the new info replace the old info)
• One way attitude can change is through direct experience > you could have a +ve or -ve view towards something but this could change if you have an experience challenging this attitude > often takes more than one exposure as attitude is usually quite entrenched or one very impactful experience
• Observational learning: this refers to modelling or copying parents or peers
• Classical Conditioning: association forms between two different things, depending on if it is a +ve or -ve association depends on how the association may change (e.g.: Pavlov’s dog had a +ve association with the bell because that was associated with food)
• Operant conditioning: Positive or negative reinforcement > praise and punishment reinforces certain attitudes or can change it
• Social roles, norms, culture + mass media
Way of measuring overt attitudes ^ (conscious + visible)
How do we measure attitudes?
• Attitude scales: Most common way of measuring attitude because it is very easy to give ppts but there are 3 distinct ways the attitude scales can be obtained
Measuring attitudes: Thurston (1928) approach
- Ppts sort 100s of statements (given a pack of cards or in a computer programme) into categories ranging from favourable-unfavourable. Statements with the highest level of agreement can be used to form a scale to give to other Ppts. > e.g. do all ppts put certain statements as favourable or unfavourable. The more stable these are, the better chance of giving an established scale to other ppts because you have already piloted the attitudinal scale with something existing
Evaluation: Thurston (1928) approach
• This is very time consuming and reliability isn’t guaranteed because it depends on the wording of statements + how they are interpreted and if the sample is being honest. Also relies on subjectivity because what one person sees as favourable, another may not > can be good at initial stages of research when deciding what qs to ask
Measuring attitudes: Likert (1932) approach
- (Most commonly used) Ps respond to statements based on a 5 (or 7) point scale ranging from strongly agree to strongly disagree. An attitude score is then calculated from the total based on responses which forms the attitude scale. (e.g. if we are asking about attitudes to higher ed, within those there could be sub-scales attitudes like attitude to lecture > so different questions measure different aspects of higher ed, then come up with total score for each of those). If the questions are worded in a certain way, they may have to be recoded > Some questions may need to be reverse scored (wording) so a high score always means the same thing. > in Likert scales, every score needs to represent the same thing (e.g. a low score on a question should equal a negative attitude OR positive attitude) > if statements are worded in the opposite way where a high score would give the opposite of what it should, then it has to be recoded (e.g. a score of 5 would be a score of 1 or a score of 4 should be a score of 2) > can be easy to mix up but this approach is quicker
Evaluation: Likert (1932) approach
• Researchers must beware of acquiescence response set (tendency to agree), social desirability and issues relating to wording (avoid double-barrelled qs + leading qs), measures to check for lying and attention > why we use counter wording for question so it makes the ppts think before they respond + can also use attention checks. Likert scales require a good understanding of psychology and attitude being measure already
- “bogus pipeline” (Jones & Sigall, 1971) can help avoid ppts lying before they actually do it > tricks ppts into thinking there is an extra physiological measure like a lie detector test at the same time of them doing the qs so the ppts won’t lie because they think the researcher will know > need strong grounds to do this because it is deception
Measuring attitudes: Guttman (1944)
- argued that neither of these approaches capture unique meaning because people can obtain the same overall score but actually agree and disagree on different qs. Guttman suggested measuring a single unidimensional trait instead > Statements are ordered along a continuum ranging from least to most extreme. P’s indicate the most extreme statement they accept. It is then assumed that participant accept all items which are less extreme than this one.
Evaluation: Guttman (1944)
- very difficult to design a reliable and valid scale which captures this full range and it assumes that someone holds an attitude to begin with. > low reliability because what could be an extreme attitude to one person may not be for another person. Captures the unique element but is not as generalisable + difficult to capture the full range because it relies on the researcher knowing what is most extreme and least extreme > subjective. (extreme attitudes could be underground + not well known) + also assumes the person holds an attitude towards the thing they are measuring (they could be neutral + not socially significant)
Measures for implicit attitudes
- need to know of implicit measures as people aren’t always aware of the attitude they hold (unconscious)
- Bias in language: People tend to use more concrete language when speaking about socially desirable attitudes and more abstract language when speaking about socially undesirable attitudes (Maass, 1999). Discourse analysis where you look deep into the text to see what is coming out + can be used to explore attitudes in transcripts.
• Look for things like hesitation when discussing certain ideas/attitudes, can look at different speech
• qualitative approach and allows for open questions to be used so it is especially useful when exploring sensitive topics because it allows people to talk around topics + in their own time w/o being forced
• Attitude Priming (and Implicit Association Tasks): We tend to make judgements quicker if they are consistent with our own attitude e.g. stimuli are presented on screen and Ps are asked to indicate if it was ‘good’ or ‘bad’. Reaction times are recorded. Faster responses = consistent with own attitude, longer = inconsistent. This is often used to explore socially undesirable or controversial attitudes.
• However, a longer response could mean other things too (e.g. distraction or difficulty understanding, reduction in attention).
Attitudes + health
If there are unhealthy behaviours like alcohol use, it is generally assumed that attitude towards these things is influencing behaviour, so to change the behaviour we need to change the attitude > other factors also have an impact on behaviour too such as age, gender and culture
Why use Health Behaviour Models?
- To explain, predict, and change attitudes & behaviour because we can see different elements which are theorized to impact behaviour and try change these elements of the attitudes to bring about more positive behaviour
- Offer hypotheses about the influence the attitudes can have on behaviour > helpful in designing research studies thus help develop effective intervention to help reduce harmful behaviour + promote healthy behaviour
- The extent to which attitudes can predict behaviour varies considerably and a notable reason for this is how attitude measures are designed. The wording of the questions in attitude scales (and how they are combined to calculate a total attitude score) can make a big difference in how well the attitude predicts behaviour (Ajzen & Fishbein, 1975; Davidson & Jacard, 1979). > can combine different questions to create different attitude scales + the more specific the scale is the better it is as predicting behaviour
Predictive power can vary: Specificity
• Davidson and Jacard (1979) investigated how the specificity of their attitudinal measure about birth control effected their ability to predict women’s use of birth control pills.
• Had lots of qs which were combined in different ways to try and identify how specific an attitude scale needed to be to be the best predictor of behaviour
• Attitudes scores were correlated against actual use:
1. Questions about Attitude towards birth control (r = 0.08). Far too broad > not telling us about what kind of birth control, is it personal use or other people’s use > low +ve corr between attitude scale and behaviour (using birth control pills)
2. Attitude towards birth control pills (r = 0.32). Still too broad but getting better > asking about a certain kind of birth control which increased the correlation between attitude and behaviour. Slightly stronger correlation between attitude + birth control
3. Attitude towards using birth control pills (r = 0.53). This has now become more specific to the type of birth control and personal use but not timescales. Stronger correlation
4. Attitude towards using birth control pills during the next two years (r = 0.57). This is now specific to the type of birth control, personal use and timeframe
• The more specific the attitudinal measure was, the better it predicts use of birth control pills. However, more general attitudes can also predict behaviour if they look at a collection of behaviours instead of one specific behaviour (e.g. being environmentally conscious tends to predict a variety of environmentally-friendly behaviours such as diet + recycling habits).
