Class Notes (838,386)
Canada (510,872)
Psychology (6,266)

Attitude measurement.docx

4 Pages
Unlock Document

Psychology 3723F/G
Martin Kavaliers

Attitude measurement Attitudes are latent constructs and thus have to be inferred from indirect information (observed variables) Reliability: test-retest reliability, internal reliability Validity: face, content, convergent/discriminant, predictive Most measures rely on self-report  most convenient  people are aware of explicit attitudes (aka not implicit – diff techs)  people should be willing to report explicit attitudes honestly o unless strong social desirability demands, when some attitudes are regarded as undesirable – require diff techniques  thurstone’s method of equal-appearing intervals o goal: to create scale w “interval” properties: each unit represents the same increase in favourability o ask judges to sort 15-20 statements into 11 “equally spaced” piles o select limited number of items (7-9) w approx. equally spaced scale values o take the mean of values to get “attitude score” o problems:  time-consuming to create  interval property of scores is doubtful  likert’s method of summated ratings o simpler o generate pool of clearly un/favorable statements on issue – no neutral o initial group indicate dis/agreement w each item on a varying “7” point scale i.e. strongly disagree – strongly agree  all items scored so that high scores rep same attitude  conduct “item analysis” to eliminate ambiguous/invalid items - i.e. item-total correlation must exceed .3 o final scale includes only items that passed item analysis o problems:  time-consuming to create  agree-disagree ratings can be ambiguous  osgood’s semantic differential scale o simpler than both o designed to measure “connotative meaning” of objects/concepts o based on factor analyses of peoples’ ratings of many different objects on a wide array of bipolar adjective scales o 3 underlying dimensions were identified:  evaluation (good-bad) - represents attitudes - p’s rate attitude object on bipolar eval scales i.e.  good-bad, beneficial-harmful, wise-foolish, like-dislike (all at once) - can be used to measure attitude toward any object/issue  potency (strong-weak)  activity (active-passive) o problems:  need to make sure chosen evaluative dimensions make sense with regard to attitude object  jaccard et al. examined reliability and convergent-discriminant validity of these rating scales (plus single-item rating scales(?)) o measured attitudes toward smoking and capital punishment on two different occasions using each of the 4 techniques o high test-retest reliability for all o high convergent validity btw o low method variance o low discriminant validity  all measures of explicit attitudes to date have used “absolute” rating scales  olson, goffin, & hayes tested whether “relative” measures of attitudes might be better than absolute measures o relative measures require comparisons of target to others  i.e. taller or shorter o absolute easures reflect qualities of the target directly  i.e. height in inches o based on goffin’s earlier work on work setting performance – relative percentile method  rate performance on scale from 0-100, # = percentage of workers whose performance is worse than target  also obtained objective measures; predicted less accurately o used relative percentile method to measure attitudes  i.e. compare yourself to all students in psych 2020 at uwo o half reported attitudes using a relative measure, half using absolute o separate questionnaire had p’s report frequency of associated behav’s  i.e. 1 questionnaire – attitude towards organized religion? 2nd questionnaire – how many times attended religious service o relative questionnaire reported more significant values – predict behaviour better – why? 1. Respondents must consider more information befor
More Less

Related notes for Psychology 3723F/G

Log In


Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.