


Study with the several resources on Docsity
Earn points by helping other students or get them with a premium plan
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
Community
Ask the community for help and clear up your study doubts
Discover the best universities in your country according to Docsity users
Free resources
Download our free guides on studying techniques, anxiety management strategies, and thesis advice from Docsity tutors
Research study designs in psychology, test-retest coefficient, validity,
Typology: Study notes
1 / 4
This page cannot be seen from the preview
Don't miss anything!
receive a score of 5 for that item. If they answer "strongly disagree" to the item "I am outgoing," then they receive a 1. So, when people receive scores close to 5 they are considered extraverted. When people receive scores closer to one they are considered low in extraversion or high in introversion. To calculate one's overall score for extraversion/introversion on the above scale, you calculate the average score across the four items. So, you add up the four scores and divide by the total number of items - 4. The average should always be in the same range as the scale. So, for the above scale the average should fall between 1 and 5. To reverse score, before we calculate the overall average, we need to first change the scoring for items that are written in the reverse or opposite direction. So, for the above scale we need to change the scoring for "I don't talk a lot" and "I keep in the background." To reverse score, you take the original rating for each item and change to the opposite score. So, if someone answered strongly disagree (a 1) for "I don't talk a lot," we change their score to a 5 to indicate they are high in extraversion. The major advantage to including reverse-scored items in a scale is to ensure participants are paying attention to the scale items and answering honestly. Disadvantages to reverse-scored items is that these items increase cognitive processing in participants. When reading items coded int he same direction, a participant can easily think about and rate their personality. But, reverse-scored items require participants to stop and think about the trait in the opposite direction. So, it is possible that reverse-scored items might be confusing to participants and they could actually decrease reliability.
When developing a self-report measure, researchers consider two main issues: reliability : Does the scale consistently measure the personality trait?
validity : Does the scale measure the personality trait accurately? Reliability is the ability of a scale to measure a personality trait consistently. Reliability scores are correlations, thus ranging from -1 through 0 to +1. A score closer to +1 indicates a high level of reliability. Personality psychologists calculate reliability using one of the following three measures: the test-retest coefficient, the parallel-forms coefficient, or the internal consistency coefficient. Test-retest coefficient measures how one sample of participants scores on the same test at two different periods of time. If you find that the test-retest coefficient is close to +1 (say, around +0.70 or above), then you would conclude that the test has high reliability. Test-retest reliability scores are higher when the interval between the two tests is shorter. As the time between the two tests increases, reliability scores become lower. A major problem with using the test- retest coefficient to measure reliability is that, at Time 2, participants may remember the questions on the exam from Time 1 and, at Time 2, answer in the way they think researchers expect. One way to account for this problem is to use two different versions of the personality scale at Time 1 and Time 2. This is called the parallel-forms coefficient. Parallel-forms coefficient is a second way to measure reliability. In this test of reliability, the researcher develops two measures of the same scale. For instance, the researcher would develop Version A and Version B of the scale for extraversion. Version A and Version B would both measure extraversion, but would ask different questions. Then, the same group of 100 participants would complete Version A on September 1 and Version B on October
Internal consistency is a third way to measure the reliability of a scale. The internal consistency coefficient determines which grouping of test items provides the highest level of
This coefficient is determined by calculating the correlation between all split-halves of a personality measure.
of extraversion, you have achieved convergent validity. Conversely, if you find that the self-report measure has a low or zero correlation with the observer-report, then you have not achieved convergent validity. These low correlations would tell you that the self-report and observer-report scales are measuring different constructs.
Four types of designs are used in personality psychology; the most common involve S-data and O- data. Each study design has its own advantages and disadvantages. Reliability and validity provide ways for personality psychologists to assess the accuracy of their personality measures. Reliability determines whether the personality measure is consistent, whereas validity determines whether the personality measure is measuring the correct construct. A personality measure can have reliability without validity, but it must have reliability to attain validity. Remember: no validity without reliability! O-data represent an important method within personality psychology. The more observers that researchers use, the more accuracy they achieve in assessing personality. Typically, researchers aggregate observer-report scores and self-report scores to attain the best measure of a participant’s personality.