Help
Options
Didn't know it?
click below
click below
Knew it?
click below
click below
Don't Know
Remaining cards (0)
Know
retry
shuffle
restart
0:00
Embed Code - If you would like this activity on your web page, copy the script below and paste it into your web page.
Normal Size Small Size show me how
Normal Size Small Size show me how
Prueba2-
| Question | Answer |
|---|---|
| 1. A test that measures what it is designed to measure is____; a test that produces consistent/stable scores is _____. | Validity; Reliability |
| 2. The correlation between IQ and achievement in a large sample is 60. Using the concept of the coefficient of nondetermination, the variance is not predictable from the relationship is | 64% |
| 3. What type of correlation is used calculate item-total correlation when 1 variable is dichotomous and other interval scale | point biseral |
| X=T+E, x refers to ___, E refers to ___ | person’s obtained score/measure error |
| What is accurate conclusion drawn from a reliability coefficient of | .90 |
| Generally considered to be major source of measurement error… | content sampling |
| The degree of reliaibility decreases as | proportion of observed variance is due to measurement error |
| If 8% of test scores observed variance is due to measurement error, the reliability coefficient of test would be | .92 |
| Reliability index reflects the correlation btw | observed scores and true scores |
| Test-retest reliability | primarily evaluates measurement error due to temporal instability. |
| 11. The mean of split-half coefficients resulting from all possible split-half correlations of a test is known as | coefficient alpha |
| Alternate form reliability with delayed administration | evaluates ME due to both content and temporal instability. |
| 13. Test-retest approach to estimating reliability is not appropriate if test is subject to | carry over effects. |
| Split-half reliability | primarily evaluates measurement error due to administrative errors |
| What is an advantage of internal consistency reliability estimates | can be calculated from one administration of test. |
| 16. Coefficient alpha produces reliability estimates____ those produced by the split half method. | smaller |
| Alternate form reliability with simultaneous administration | primarily evaluates due to content sampling. |
| ___takes into account item homogeneity. | coefficient alpha. |
| 19. Which of the following is an appropriate estimate of reliability when one is assessing the reliability of a highly speeded test | test-retest reliability |
| Spearman-Brown formula is appropriate used to | correct split haft coefficient |
| 21. Reliability of composite scores is generally____ the reliability of the individual tests. Reliability of difference scores is generally ____the reliability of the individual tests. | greater than; less than. |
| As reliability decreases, the range within which we expect true scores to fall (i.e. confidence intervals | increases |
| 23. If the standard error of measurement of a test decreases the range within which we expect true scores to fall | decreases |
| If a 3rd grade math test ….if the test were to intended only to measure, the inclusion of reading comprehension would reflect | construct irrelevant variance. |
| 25. If a predictor test correlates highly with a criterion measure, test scores can be used to predict criterion scores through | Linear regression |
| ___produced by measurement of stable characteristic other than those the test is designed to measure. | systematic measurement error |
| 28. Math test to cover year’s math curriculum administered. Teacher ran out of time and only asked first semester’s material….what type of validity threat is? | construct underrepresentation |
| Item total correlation is .5?true of following interpretations | .25 |
| Relationship btw validity coefficient and standard error or estimate? | as validity coefficient increases; the standard error of estimate decreases |
| 31. Relative to ___design, the ____ design is likely to result in a more restricted rage of predicted scores | concurrent validity; predictive validity |
| 32. ___result of transient evens such as how the test taker feels that day and environment within which the test is taken on that day | random measurement error. |
| 33. Examination of relationship between the test and external variables thought to be direct measure of the construct is example of what type of validity evidence | validity based on relations to other variables |
| 34. Not appropriately classified as validity evidence based on internal structure of assessment | Evaluating how well the test items sample the behavior or subject matter being measured |
| 36. Contemporary psychometric standards emphasize…the validity of psychometric standards emphasize | is a unitary construct |
| 39. When convergent measures are used in validation studies, we expect…. | moderate to strong (positive correlations) |
| 40. When evaluating validity coefficients, the accepted standard is to require coefficients that are | none of the above |
| 41. Which of the following would not be considered validity evidence based on relations to other variables | response processes study. |
| 42. Which is true | reliability must be present in order for scores to be valid |
| 44. How can reliability be reduced by presence of many items with p values close to 0 | this results in a restricted range of scores. |
| 45. Which allows examiner to establish CI indicating which true scores are likely to fall | standard error of estimate. |
| 46. What is optimal average Item difficulty index for a quiz that has only constructed response items | .5 |
| 47. What is the approximate optimal average item difficulty for MC with four alternatives | .75 |
| 48. Correlating score on item with the total score on the test produces a measrure similar to which item statistic | item discrimination index |
| 49. A general recommendation is to use items with p values with a range of approximately … around the optimal value for the item type | .2 |
| 50. A test item with a p=.85 and D=.10 | easy; does not discriminate well |
| 51. A test item with a p=.80 and D= .35 | easy ; discriminates well |
| 52. A test item with a p=..25 and D=.15 difficult | does not discriminate well |
| 53. A test item with p=.55 and D=.11 | intermediate; does not discriminate well |
| 54. A significant advantage of item-total correlation over item-discrimation index is that | conclusions can be presented as a variance statement |
| 55. Which of the following is most controversial in the professional literature | validity evidence based on consequences of testing |
| 56. In class of 100 students, 60 of the students answer item #1 correctly and 40 answer it incorrectly. What is p value of this item?.... | .6 |
| 57. A distracter should | be selected by some examines and demonstrated negative discrimination |
| 58. An item characteristic curve is a graph on which…is plotted on the horizontal axis and …is plotted on the vertical axis | the probability of a correct response; ability |
| 59. For item#1, pT is .70, pB is .2. what is the value of D for this item | +.50 |
| 60. On speed tests, measures of item difficulty and discrimination largely reflect | location of the item in the test. |
Created by:
44409659