VIEWS: 12 PAGES: 5 POSTED ON: 8/10/2011
GRE Questions related to basic inferential statistics and probability: 1. The probability that the null hypothesis (H0) will be rejected when it is in fact true is called a(n): A. type II error. B. type I error. C. two-tailed test. D. experimental error. E. false validity 2. A psychologist wants to determine whether one group of 15-year-old girls’ I.Q. scores differs significantly from that of a second group of 15-year-old girls. He would use: A. a t-test for two independent means. B. a t-test for sample and population means. C. chi-square D. a factorial design E. the Pearson product-moment correlation coefficient. 3. When the difference between two means is shown to be significant: A. the null hypothesis is rejected. B. the null hypothesis is disproved. C. the alternative hypothesis is disproved. D. the independent variable is proved. E. both A and B. 4. The Z-table is derived from: A. the standard deviation B. random sampling. C. the normal distribution D. the z-scores. E. the square root of the t-scores. 5. Which of the following choices best describes the correlation pictured above? A. correlation = +1.00 B. correlation = -1.00 C. correlation = 0.00 D. correlation = -0.45 E. correlation = +0.45 6. Which of the following correlation values is the best predictor for a relationship between x and y? A. –7.0 B. +.60 C. +.50 D. .10 E. +7.0 7. Which of the following correlation coefficients represents the strongest relation between two variables? A. .01 B. -.28 C. .38 D. .58 E. -.68 8. Which correlation coefficient below has the most statistical significance? A. +.27 B. 0.0 C. -.34 D. +.82 E. -.96 1 9. In a statistical test, Power (1 - ) is: A. the probability of accepting a false null hypothesis. B. a measure of external and construct validity C. the probability of rejecting a false null hypothesis D. a type I error. E. a type II error. 10. Which of the following concepts concerns descriptive statistics? A. Chi square B. Analysis of variance. C. T-test. D. F test E. Mode 11. A frequency-distribution will often approach the normal distribution as: A. the number of scores included gets very large. B. the number of scores included gets very small C. more variables are included in the frequency distribution. D. certain scores are eliminated from the distribution. E. the distribution normalizes in shape. 12. Given a sample mean, the range of values within which the parametric mean must lie, can be computed. This is known as the: A. parametric deviation B. variability range. C. mode. D. confidence interval. E. parametric range. 13. Which of the following statements is correct? A. A factorial design is not based on analysis of variance. B. Analysis of covariance is not related to correlation. C. Correlation is related to predictability in methodology. D. The range of correlation coefficients is -.04 to +.04. E. both B and C 14. In factor analysis, separateness of factors is assessed by: A. degree of correlation. B. task analysis C. similarity of distribution D. A and B E. B and C 15. A significant correlation means that: A. a high correlation is more likely to exist between significant variables. B. the likelihood of getting a correlation that large is very probable. C. a relationship is likely to exist between two variables in the general population. D. none of the above. E. A and C 16. The number of independent pieces of information remaining following an estimation of population parameters is the: A. degrees of freedom (df). B. error term. C. critical value (F) D. data reduction. E. Scheffe critical value (Fs). 17. The probability of committing a type I error is defined as: A. beta () B. mu () C. 2 (chi) D. 2 A (omega) E. alpha () 2 18. A Z-distribution refers to: A. a measure of variability; an average of the sum of the squares deviations from the mean. B. coefficients used in calculating the sum of squares for single-df comparisons. C. the normal distribution expressed in terms of deviations from the mean divided by the standard deviation. D. a theoretical distribution used in conjunction with the t-test and in the establishment of a confidence interval. E. a distribution always used in conjunction with the chi-square test of correlation. 19. The mean of the squared differences from the mean of the distribution is a definition of the: A. Z-score B. mode. C. median. D. standard deviation. E. variance The following question refers to the following paragraph: A researcher has tested a new drug for reducing anxiety on a sample of 100 people. After a treatment period of two weeks, 70 people reported feeling much less anxious and certain physiological tests showed them to be less anxious. The researcher now wants to claim that the drug is effective. 20. If the researcher rejects the null hypothesis that the drug is not effective but in reality the null hypothesis is true, he has: A. increased the power of the test. B. made a Type I error. C. Made a Type II error. D. increased E. none of the above. 21. A newly developed children’s intelligence test is reported in the literature to be correlated +.25 with the Wechsler Intelligence Scale for Children – Revised from (WISC-R). Which of the following statements provide the best assessment of the above information? A. The new intelligence test has high criterion validity B. The new intelligence test is relatively content-independent of the WISC-R. C. The new intelligence test has low concurrent validity. D. The new intelligence test has high reliability. E. The new intelligence test has low reliability. 22. Which of the following is true? A. 33% of the scores fall between the median and 3rd quartile. B. 50% of the scores fall between the median and 3rd quartile. C. 25% of the scores fall between the median and 3rd quartile. D. 66% of the scores fall between the median and 3rd quartile. E. 75% of the scores fall between the median and 3rd quartile. The following two questions refer to the following passage. To use the F-table, you would need information about the body of data which you are analyzing. It is important to know the degrees of freedom for the numerator mean square of the F-ratio. You also need to know the denominator degrees of freedom for the F ratio. Finally it is important to know the significance level (). All of this knowledge is necessary to successfully determine F. 23. The degrees of freedom (df) represent: A. the difference between a score and a mean of the set of data scores. B. an error term. C. the number of independent pieces of information. D. a value obtained for a sampling distribution that specifies a critical region. E. the denominator term of an F ratio. 24. In an Analysis of Variance, the significance level indicates: A. the probability of making a Type I error. B. that F is correct. C. the probability of making a Type II error. 3 D. A and B E. B and C The following two questions refer to the following graph: 25. What is the Z-score for the obtained raw score, 64? _______ 26. What percent of the values in this distribution fall between 54 and 66? _________ The following three questions refer to the following passage: The power of a statistical test is the probability of correctly rejecting the null hypothesis when it is false. It is defined by the value 1- where (beta) is the probability of not rejecting the null hypothesis when it is false. In testing the null hypothesis, it is compared to alternative conditions. If this alternative hypothesis specified direction, a one-tailed test is utilized. If the direction of the alternative values is not specified, a two-tailed test is used. 27. If power is increased by changing the critical level, , from .01 to .05 then the probability of making a: A. Type II error increases. B. Type II error decreases. C. Type I error increases. D. Both B and C E. Both A and B 28. Using a one-tailed test will: A. always increase the power. B. increase the power when the true mean is in the direction of the tail. C. always increase the probability of rejecting the null hypothesis. D. both A and B E. never increase the power. 29. In order to increase the power of a test, a researcher can: A. decrease the sample size. B. increase C. increase the sample size. D. both B and C E. both A and B 30. A test score that has not been converted into a form permitting comparison with scores of other tests is known as a: A. rank score B. percentile score. C. raw score. D. T-score. E. Z-score. 31. If the correlation coefficient between two variables equals .98, what can you conclude? A. the variables are causally related. B. there is relatively little unexplained error variance. C. as one variable increases in value the other decreases. D. the means of the variables are equal. E. the standard deviations of the variables are large, close to 1.0. 32. Which of the following is needed in computing a t-test? A. Median B. Standard deviation C. Correlational coefficient. D. Chi square E. Interquartile range 33. Suppose you are told that two conditions in an experiment differ at the .05 level of significance. What does this imply? 4 A. the difference occurs only 5% of the time. B. the difference can be generalized to 5% of the population. C. the difference is not very reliable. D. the difference is probably correlational, not causal E. the difference occurs by chance only 5% of the time. 34. Suppose that two treatment groups produce frequency distributions that do not overlap. What can you conclude? A. The difference in means of the two groups occurs by chance less than one time in a hundred. B. The difference in means of the two groups is less than their standard deviations. C. The difference in the standard deviations of the two groups is negligible. D. The difference in the standard deviations of the two groups is likely to occur less than one time in a hundred. E. The difference in means of the two groups is most likely due to chance. 35. If two variables are correlated and yield an r value equal to -.75, then which of the following statements is true? A. The negative value implies a chance relationship between the variables. B. About 50% of the variance in the scores of one variable can be explained in terms of the other variable. C. The standardized mean of one variable differs from the standardized mean of the other variable by .75. D. The probability that the correlation occurred by chance is less than .001. E. An analysis of variance on the data would reveal a significant causal relationship. 36. A psychologist is studying the relationship is studying the relationship between verbal learning and mode of presentation. Upon analyzing the data, the psychologist finds a correlation of +1.50. On the basis of this correlation, he would conclude that there is a: A. strong positive correlation. B. strong negative correlation. C. computational error. D. low negative correlation. E. low positive correlation. 37. Any procedural variable which can cause a subset of the population to be non-representative of the population is a: A. sample error. B. sample bias. C. population error. D. sample shift. E. population bias. 38. Which of the following best describes the alternative hypothesis? A. the hypothesis that there is no effect. B. the hypothesis that there is only one effect. C. the hypothesis that there are two effects. D. the null hypothesis is false. E. none of these 39. Which of the following best describes correlational analysis? A. a measure of association between two variables. B. a measure of linear association between two characteristics C. a measure of causation between X and Y. D. a measure of causation of X on Y. E. a measure of causation of Y on X. 40. When the difference between two means is shown to be significant: F. the null hypothesis is rejected. G. the null hypothesis is disproved. H. the alternative hypothesis is disproved. I. the independent variable is proved. J. both A and B. 5