In: Statistics and Probability
A.
College Students’ Recent Life Experiences. Researchers administered to 216 undergraduate students (in the same time period) the Inventory of College Students’ Experiences and a measure of daily hassles. The total coefficient alpha was .92 for the Inventory and .96 for the measure of daily hassles. The Inventory correlated with daily hassles at .76 (p <. 001).
B.
Test scores for the Eysenck Personality Inventory and Cattell’s 16 Personality Factor Questionnaire (16PF) were obtained for male army applicants for flyer training. Forms A and B were used for each test and the correlations between forms for the same test ranged from .39 to .85. Some of the men entered flying school several years after taking the tests. The correlations of the subscales on the two tests with training outcome (pass or fail) averaged about .20.
C.
Computer Aptitude and Computer Anxiety. Researchers gave 162 students enrolled in computer courses a test that measured computer anxiety and another that measured computer aptitude. Both were given at the beginning of the course. Student performance in the course was measured by the grades they earned in the course. Computer aptitude was correlated with course grade at .41 (p <. 01) for one course and .13 (p < ns) for the other. Correlations of computer anxiety and course grade were .01 and .16 (p < ns).
Identify the following for each of the studies above:
Predictor(s)
Criterion
Type of validity estimated
Validity Coefficient
Reliability Coefficient
ANSWER:
Given that,
A)
The tricky part is that a test can be reliable without being valid. However, a test cannot be valid unless it is reliable. An assessment can provide you with consistent results, making it reliable, but unless it is measuring what you are supposed to measure, it is not valid.
Therefore the above test is validity rest...
B)
1)
The biggest confounding variable is the inclusion of the population who entered the flying school several years after taking the test. this may have caused the reliability of the tests to go down.
2)
the obtained correlation of .20 is very weak positive correlation, therefore to establish the success of the training based on this number is not possible.
3)
The reliability score of .39 to .85 between the two personality questionnaires ranges from weak positive to strong positive. This is bound to happen as both the personality questionnaires are based on the factor analysis model of personality.
C)
Predictor(s) : Computer Aptitude and Computer Anxiety.
Criteria : 162 students enrolled in computer courses.
Type of evidence (concurrent/predictive) : Predictive i.e. Grades
Validity coefficient and its strength :
Computer aptitude was correlated with course grade at 0.41 (p <0. 01) for one course
Strength = r2 = (0.41)*(0.41) = 0.1681*100 = 16.81%
Computer aptitude was correlated with course grade at 0.13 (p <ns) for another course
Strength = r2 = (0.13)*(0.13) = 0.0169 *100= 1.69%