1 / 51

Measurement of Attitudes

Measurement of Attitudes. Dr. K. A. Korb University of Jos. Overview.

Télécharger la présentation

Measurement of Attitudes

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Measurement of Attitudes Dr. K. A. Korb University of Jos

  2. Overview A majority of Americans favor having Arabs, even those who are U.S. citizens, being subjected to separate, more intensive security procedures at airports. About half of Americans favor requiring Arabs, even those who are citizens of the United States, to carry special ID. Gallup.org

  3. Dr. K. A. Korb University of Jos Overview(Mike Frone) • Please tell me if you would favor or oppose each of the following as a means of preventing terrorist attacks in the United States. • Requiring Arabs, including those who are U.S. citizens, to undergo special, more intensive security checks before boarding airplanes in the U.S. • Requiring Arabs, including those who are U.S. citizens, to carry a special ID. • Problems • Leading stem • Double-barreled questions

  4. Dr. K. A. Korb University of Jos Overview • Considerations in Measurement • Types of Attitude Measurement • Explicit • Likert Scale • Interview • Implicit • Implicit Association Test

  5. Dr. K. A. Korb University of Jos Reliability: Consistency of results Reliable Reliable Unreliable

  6. Dr. K. A. Korb University of Jos Reliability Theory • Actual score on test = True score + Error • True Score: Hypothetical actual score on test • The more reliable the test is, the less error in measurement

  7. Dr. K. A. Korb University of Jos Reliability: Sources of Error • Error in Test Construction • Error in Item Sampling: Items measure more than one construct • Error in Test Administration • Test environment • Test-taker variables • Examiner-related variables • Error in Test Scoring

  8. Dr. K. A. Korb University of Jos Validity: Measuring what is supposed to be measured Valid Invalid Invalid

  9. Dr. K. A. Korb University of Jos Validity • Three types of validity necessary for all psychological tests: • Construct validity: Measure the appropriate psychological construct • Criterion validity: Predict appropriate outcomes • Content validity: Adequate sample of content

  10. Dr. K. A. Korb University of Jos Construct Validity • Definition: Appropriateness of inferences drawn from test scores regarding an individual’s status of the psychological construct of interest • Two considerations: • Construct underrepresentation • Construct irrelevant variance

  11. Dr. K. A. Korb University of Jos Construct Validity • Construct underrepresentation: A test does not measure all important aspects of the construct. • Construct-irrelevant variance: Test scores are affected by other unrelated processes

  12. Dr. K. A. Korb University of Jos Sources of Construct Validity Evidence • Homogeneity: The test measures a single construct • Evidence: Good reliability • Convergence: Test is related to other measures of the same construct and related constructs • Evidence: Highly correlations with other measures

  13. Dr. K. A. Korb University of Jos Sources of Construct Validity Evidence • Theory: The test behaves according to theoretical propositions about the construct • Evidence by changes in test scores according to age: Scores on the measure should change by age as predicted by theory. • Evidence from treatments: Scores on the measure change as predicted by theory from a treatment between pretest and posttest.

  14. Dr. K. A. Korb University of Jos Criterion Validity • Definition: Correlation between the measure and a criterion. • Criterion: Other accepted measures of the construct or measures of other constructs similar in nature. • A criterion can consist of any standard with which the test should be related • Examples: • Behavior • Judgments • Other instruments related to attitudes

  15. Dr. K. A. Korb University of Jos Criterion Validity • Three types: • Convergent validity: High correlations with measures of similar constructs taken at the same time. • Divergent validity: Low correlations with measures of different constructs taken at the same time. • Predictive validity: High correlation with a criterion in the future

  16. Dr. K. A. Korb University of Jos Criterion Validity Example High correlations with other minerals indicates good criterion validity. Low correlations with unrelated beverages indicates good criterion validity.

  17. Dr. K. A. Korb University of Jos Content Validity • Definition: Sampling the entire domain of the construct

  18. Dr. K. A. Korb University of Jos Measuring Attitude Overview • Three Methods to Assess Attitudes • Explicit (Self Report) • Questionnaire • Interview • Implicit • Behavioral Measures • Physiological Response • Reaction Time

  19. Dr. K. A. Korb University of Jos Scaling of Attitudes • Types of Scales • Semantic Differential Scales • Reactions to stimulus on a bipolar scale with opposite adjectives at the end • Thurstone Scaling • Use expert judges to create an ordered list of attitudes toward a topic • Likert Scaling • Evaluation of individual items according to level of agreement or disagreement

  20. Dr. K. A. Korb University of Jos Semantic Differential Scales • Tick your opinion of UniJos: Very Bad _:_:_:_:_:_:_ Very Good Very Weak _:_:_:_:_:_:_ Very Strong 0 -3 +3 0 -3 +3

  21. Dr. K. A. Korb University of Jos Thurstone Scale • Tick if you agree with the item. • People with AIDS should be considered the lowest class of human beings. • People with AIDS must be kept apart in social affairs. • I am not interested in how people with AIDS rate socially. • A refusal to accept a person with AIDS is a prejudice which should be overcome. • I believe that people with AIDS deserve the same social privileges.

  22. Dr. K. A. Korb University of Jos Likert Scaling • Define the attitude to be measured • Develop about 25 items thought to measure the attitude • Participants indicate the degree of agreement to each item • Single items work for very specific attitudes • Multiple items have the best reliability and validity for general attitudes (at least 10) • Level of Measurement: 4, 5, 7, 10

  23. Dr. K. A. Korb University of Jos Likert Scaling

  24. Dr. K. A. Korb University of Jos Likert Scaling • Factor Analyze the items to determine whether they measure the attitude • Factor Loading tells how well the item relates to the overall score • Interpret Factor Loadings as a correlation • Calculate the Reliability of the final items to establish internal consistency • Coefficient Alpha > .70 • Analyze • Sum responses to each attitude construct separately • Average responses to each attitude construct separately

  25. Dr. K. A. Korb University of Jos Likert Scaling

  26. Dr. K. A. Korb University of Jos Likert Scaling

  27. Dr. K. A. Korb University of Jos Likert Scaling Summed Score = 2+3+1 = 6 Averaged Score = 2+3+1 = 6 6 3 = 2

  28. Dr. K. A. Korb University of Jos Likert Scales • Potential Problems • Acquiescence bias: Agree with statements as presented • Solution: Develop equal numbers of positive and negative items • Reverse score negative items • Central tendency bias: Avoid extreme categories • Strongly Agree, Strongly Disagree • Social desirability: Tendency to portray oneself in favorable light • Solution: Bogus Pipeline

  29. Dr. K. A. Korb University of Jos Acquiescence Bias

  30. Dr. K. A. Korb University of Jos Central Tendency Bias

  31. Dr. K. A. Korb University of Jos Social Desirability

  32. Dr. K. A. Korb University of Jos Interviews • Define the purpose of the interview • Select an interview format • Standardized Open-Ended Interview: Specific set of questions with open-ended response • Semi-structured: Structured questions followed by open-ended questions to probe deeper • Structured Interview: Participants select pre-determined responses

  33. Dr. K. A. Korb University of Jos Standardized Open-Ended Interview • What issues do you think the Plateau State governor needs to spend more effort on? • How effective do you think the new road construction plans are? • How well do you think the public schools are being managed in Plateau State?

  34. Dr. K. A. Korb University of Jos Semi-Structured Interview • What teacher has most influenced you? Why? • She encouraged me to pursue my dreams. • How did she encourage you to pursue your dreams?

  35. Dr. K. A. Korb University of Jos Structured Interview • How well do you think that Plateau State governor compares to other Nigerian governors in providing funds for education? • Better than most Nigerian governors • Same as most Nigerian governors • Worse than most Nigerian governors • How well do you think that Plateau State governor compares to other Nigerian governors in fixing the roads? • How well do you think that Plateau State governor compares to other Nigerian governors in providing jobs?

  36. Dr. K. A. Korb University of Jos Interviews • Data Analysis • Tally responses for Structured Interview • Open-Ended questions require a categorization system • Cared for me • Encouraged to excellence • Knew the course material

  37. Dr. K. A. Korb University of Jos Questionnaires vs. Interviews • Questionnaire Advantages • Sample more participants • Quicker • Standardized • Anonymous • Interview Advantages • Probe deeper • Adaptability • Use with illiterate participants

  38. Dr. K. A. Korb University of Jos Quality of Attitude Items(Trochim, 2006) • Is the question useful? • Only ask questions that will be used for analysis • Are several questions needed? • Double-barreled question: Question has two parts • Plateau State should give more money to education and road upkeep. • I do not like pounded yam because it tastes slimy.

  39. Dr. K. A. Korb University of Jos Quality of Attitude Items(Trochim, 2006) • Will all respondents have the needed information to answer the question? • Knowledge • Vocabulary • Does the question need to be more specific? • Is the question biased? • Question leads to a particular response

  40. Dr. K. A. Korb University of Jos Quality of Attitude Items(Trochim, 2006) • Will the respondent answer truthfully? • Social Desirability • Bias in Questioning • Can the question be misunderstood? • Conclusion: Read each question as if you were the respondent. How could a respondent misunderstand the question?

  41. Dr. K. A. Korb University of Jos Implicit Measures • Purpose: Measure unconscious thoughts and feelings • Assumptions • People may be unwilling to report their true attitudes • People may not know their attitudes • Implicit Attitudes: Attitudes that people are either unwilling or unable to report

  42. Dr. K. A. Korb University of Jos Implicit Measures • Behavioral Measures • Distance from another person • Physiological Measures • Facial EMG: Measuring imperceptible muscle movements in the face • EEG: Brain wave patterns • Reaction Time • Implicit Priming Procedure

  43. Dr. K. A. Korb University of Jos Implicit Priming

  44. Dr. K. A. Korb University of Jos Implicit Priming Good Bad Pleasant Lovely Awful Nice

  45. Dr. K. A. Korb University of Jos Implicit Priming Good Boys Bad Girls Pleasant John Martha Nice

  46. Dr. K. A. Korb University of Jos Implicit Priming Good Girls Bad Boys Pleasant John Martha Nice

  47. Dr. K. A. Korb University of Jos Implicit Measures • Implicit Association Test • https://implicit.harvard.edu/implicit/demo/selectatest.html • Gender-Career • Skin Tone • Disability • Weight • Religion

  48. Dr. K. A. Korb University of Jos Implicit Measures • Race IAT Predicts: • Time spent speaking: .51 • Smiling: .39 • Social comments: .32 • Potential Uses • Research • Diversity training • Misuses • Decisions of placement

  49. Dr. K. A. Korb University of Jos Explicit vs. Implicit Measures • Meta-analysis correlated attitudes to criterion measures • Implicit measures were best predictors of stereotyping and prejudice (r = .25 vs .13) • Explicit measures were best predictors for products and political preferences (r = .67 vs .41)

  50. Dr. K. A. Korb University of Jos Choosing a Measure • Research Questions/Hypotheses • Time and Resources Available • Participants

More Related