stats final
21. On a midterm exam, your score expressed as an SAT score is 550. The mean of the exam is 74, and the standard deviation is 6. Your raw score is: A) 68. B) 71. C) 77. D) 80
77
15. Which of the following leads to a larger value for the Studentized range statistic (q)? A) Larger sample sizes B) A larger number of groups C) A larger value for alpha D) All of the above
B
12. The variance of a set of z scores is equal to: A) -1.00. B) 0. C) +1.00. D) +10.0.
C
42. A decile divides the total number of cases into how many equal parts? A) 2 B) 4 C) 10 D) 100
10
43. The median divides the total number of cases into how many equal parts? A) 2 B) 4 C) 10 D) 100
2
40. In a grouped frequency distribution, the lower real limit of the 24-26 interval is: A) 23.0 B) 23.5 C) 23.9 D) 24.0
23.5
1. If ΣX = 60 and N = 20,X is equal to: A) 0.33 B) 3.0 C) 33.0 D) cannot be determined from the information given
3
41. A quartile divides the total number of cases into how many equal parts? A) 2 B) 4 C) 10 D) 100
4
2. If ΣY = 22 and N = 5,Y is equal to: A) 0.2 B) 4.4 C) 22.0 D) cannot be determined from the information given
4.4
22. On a midterm exam, your score expressed as an SAT score is 300. The mean of the exam is 70, and the standard deviation is 8. Your raw score is: A) 54. B) 58. C) 62. D) 78.
54
51. One hundred students take a 50-point test on which Richard scores 34. If 20 students score higher than 34, one other student scores 34, and everyone else scores below 34, what is Richard's percentile rank? A) 20 B) 34 C) 68 D) 80
80
47. Which of the following represents the best performance? A) The third quartile B) The 80th percentile C) The median D) The sixth decile
80th percentile
17. Using which of the following criteria of significance is most likely to lead to a Type I error? A) .10 B) .05 C) .01 D) .001
A
4. Which of the following represents the lowest score? A) An SAT score of 360 B) A T score of 41 C) A z score of -0.57 D) They are all equal
A
1. Imagine a study in which there are two repeated conditions, and both the matched t and repeated measures ANOVA are calculated. Which of the following will be true? A) The critical F will be the same as the critical t. B) The p values will be the same for both tests. C) The RM ANOVA will have a greater chance of statistical significance than the matched t. D) The matched t will have a greater chance of statistical significance than the RM ANOVA.
B
10. Thirty participants are measured before and after some treatment, and again at follow-up. The total sum of squares is found to be 1,108, the SS for the treatment equals 118, and the SS for the subjects factor equals 538. What is the F ratio for the one-way RM ANOVA? A) 3.18 B) 6.36 C) 7.57 D) 7.79
C
1. A researcher who wants to study the relationship of two or more independent variables to a dependent variable should consider using: A) multiple comparisons. B) the t test for the difference between two means. C) one-way analysis of variance. D) two-way analysis of variance.
D
1. When drawing inferences about the mean of one population, the correct statistical model to use is: A) the normal curve. B) the t distributions. C) the t distributions if the population standard deviation is known and the normal curve if the population standard deviation is not known. D) the normal curve if the population standard deviation is known and the t distributions if the population standard deviation is not known.
D
10. When using the Pearson r as an inferential statistic, the degrees of freedom are equal to: A) the number of scores minus one. B) the number of scores minus two. C) the number of pairs minus one. D) the number of pairs minus two.
D
11. If two of the pairwise comparisons following an ANOVA exceed Fisher's LSD, how many would exceed Tukey's HSD? A) One or none B) Two C) At least two D) No more than two
D
54. When we wish to estimate the standard deviation of a population based on data obtained from a sample, before taking the square root, we divide the sum of squares by: A) N B) N - 1 C) N - 2 D) None of the above
N-1
1. If we draw 1,000 samples of size 100 from a population and compute the mean of each sample, the distribution of sample means will tend to be: A) approximately normal in shape. B) approximately normal only if the raw scores in the population are normally distributed. C) negatively skewed. D) positively skewed.
approximately normal in shape
5. A histogram displays data by using: A) bars. B) cumulative frequencies. C) points on a graph that are connected. D) "stems" and "leaves."
bars
10. A frequency distribution is skewed if it: A) has only one peak. B) has at least two peaks. C) can be divided into two mirror-image halves. D) has a long "tail" at one end.
has a long "tail" at one end
15. Using the .01, rather than the .05, criterion of significance has which effect on the probability of making a Type I error? A) It gets smaller. B) It gets larger. C) It stays the same. D) It cannot be determined from the information given.
it gets smaller
47. The absolute value of a score is defined as: A) its distance from the mean. B) its distance from the median. C) the square root of the score. D) its numerical value ignoring the sign.
its numerical value ignoring the sign
35. A distribution can have more than one: A) mode. B) median. C) mean. D) mode, median, or mean.
mode
14. A very easy exam would be likely to produce a distribution that is: A) negatively skewed. B) positively skewed. C) unimodal and symmetric. D) rectangular.
negatively skewed
37. The score at or below which a given percent of the cases fall is called a: A) percentile. B) percentile rank. C) transformed score. D) z score.
percentile
35. A student obtains a score of 76 on a 100-point midterm exam. The score of 76 is called a: A) percentile. B) percentile rank. C) raw score. D) transformed score.
raw score
46. The deviation of a single score is defined as its difference from: A) every other score. B) the mean. C) the median. D) none of the above
the mean
52. If we divide the sum of squares by the number of scores minus one, we obtain the: A) unbiased variance. B) biased variance. C) standard deviation. D) none of the above.
unbiased variance
3. As the size of the sample increases, the standard error of the mean: A) becomes larger. B) becomes smaller. C) remains the same. D) cannot be determined from the information given.
becomes smaller
1. Which of the following can be expected to be positively correlated? A) Scores on a final exam and the average of all other tests in that course B) Scores on a midterm exam and a final exam in the same course C) Both of the above D) Neither of the above
both
32. Which measure is the best one to use when the data are highly skewed and the purpose is purely descriptive? A) The mean B) The median C) The mode D) None of the above
the median
38. Suppose that 10 people have exactly the same score. All of them received a score of 4.0. The variability of these 10 scores is: A) zero. B) 1.0. C) 4.0. D) Cannot be determined from the information given
zero
8. If we were able to measure the entire population, the standard error of the mean would be: A) zero. B) 1.0. C) plus infinity. D) equal to the standard deviation of the raw scores.
zero
18. On a midterm exam, your score expressed as a z score is -1.50. The mean of the exam is 70, and the standard deviation is 8. Your raw score is: A) 54. B) 58. C) 82. D) 86.
58
20. On a midterm exam, your score expressed as a T score is 40. The mean of the exam is 75, and the standard deviation is 7. Your raw score is: A) 61. B) 68. C) 82. D) 89.
68
17. On a midterm exam, your score expressed as a z score is +0.50. The mean of the exam is 65, and the standard deviation is 10. Your raw score is: A) 60. B) 65. C) 70. D) 75.
70
74. If you know that the 25th percentile of a distribution is 17 and the 75th percentile is 33, what is the value for the semi-interquartile range? A) 8 B) 16 C) 17 D) Cannot be determined from the information given
8
10. When conducting Fisher's protected t-tests, which of the following is determined directly by the researcher? A) The alpha per comparison B) The experiment-wise alpha C) The Type II error rate D) LSD
A
16. The critical t values that are used to find a confidence interval for the population mean will get larger if: A) the sample size becomes smaller. B) the level of confidence is made smaller. C) the standard deviation becomes smaller. D) the population mean becomes larger.
A
17. In a two-way ANOVA, mean squares are obtained by dividing each sum of squares by: A) its associated degrees of freedom. B) the total number of scores. C) the number of different cells. D) the total number of scores minus one.
A
2. A two-way analysis of variance always differs from a one-way analysis of variance in which of the following ways? A) More than one null hypothesis is tested. B) The independent variable has more than two levels. C) Both of the above. D) Neither of the above.
A
2. One disadvantage of a repeated-measures ANOVA (as compared to an independent-groups ANOVA on the same data) is: A) the reduction of degrees of freedom in the error term. B) the reduction in the numerator of the F ratio. C) the reduction in the size of the error term. D) the reduction in alpha.
A
2. Using a one-way analysis of variance with three groups, a researcher obtains a statistically significant F ratio. The group means are: Group 1, 25.8; Group 2, 15.4; Group 3, 20.6. If the value of LSD equals 9.7, the researcher should conclude that which population means are different? A) Only population 1 and population 2 B) Population 1 and 2 and population 1 and 3 C) Population 1 and 2 and population 2 and 3 D) Population 1 and 2, population 1 and 3, and population 2 and 3
A
2. Which of the following are likely to be negatively correlated? A) Number of errors made on a math test and a measure of that student's math aptitude B) Scores on a final exam for one course and the same student's overall grade-point average C) Both of the above D) Neither of the above
A
20. The 95% confidence interval for the difference between two means is -3.7 to 8.5. Which of the following cannot be the 99% confidence interval for the difference between the same two means? A) -3.2 to 7.9 B) -5.6 to 11.2 C) Both of the above D) Neither of the above
A
20. Why do researchers usually not use the .1 criterion of significance? A) It would make a Type I error too likely. B) It would make a Type II error too likely. C) It would make both a Type I and a Type II error too likely. D) None of the above.
A
21. What are the practical consequences of a Type I error? A) A false positive—erroneously concluding that the researcher's theory has been supported B) A false negative—erroneously concluding that the researcher's theory has not been supported C) Either a false positive or a false negative, depending on the direction (sign) of the results
A
23. If your study consists of 100 subjects divided equally among five groups, and you have calculated that SStot = 11,101 and SSbet = 8,916, what is the value for MSw? A) 23 B) 109.25 C) 115 D) 210.7
A
24. If we retain the null hypothesis about the difference between two population means using the .05 criterion of significance, what is the probability that we have made a Type I error? A) 0 B) .05 C) .95 D) Cannot be determined from the information given
A
26. For a particular one-way ANOVA, SSbet was 23 and SSw was 69. What is the value for eta-squared? A) .25 B) .33 C) .5 D) Cannot be determined without knowing the df components
A
27. A sample of N = 100 has a mean of 42.9. The population standard deviation is known to be 7.5. Which of the following null-hypothesized values of the population mean should be rejected using the .05 criterion of significance? A) 41.1 B) 42.2 C) 44.1 D) All of the above E) None of the above
A
28. A Kruskal-Wallis H test includes four groups and a total of nine participants. If SSB = 30 and H = 12*SSB / (N (N + 1)), H is equal to: A) 4.0. B) 5.0. C) 18.0. D) 1.98.
A
28. When participants can be matched, the matched t test is preferable to the t test with independent samples because it: A) is more likely to reject a false null hypothesis. B) is more likely to reject a true null hypothesis. C) is more likely to retain the null hypothesis. D) has more degrees of freedom.
A
3. If we switch from the 95% confidence interval to the 99% confidence interval: A) we can be more sure that the population mean falls within the confidence interval. B) we will estimate the population mean more precisely. C) both of the above. D) neither of the above.
A
30. If a Kruskal-Wallis H is statistically significant, the strength of the relationship between group membership and ranks is determined by computing: A) eta squared. B) the Glass rank-biserial correlation. C) Pearson's r. D) the phi coefficient.
A
32. If a matched t test is carried out and the 95% confidence interval is computed, for which of the following intervals should the researcher retain the null hypothesis of zero change? A) -11.6 to .79 B) -23.1 to -6.8 C) Both of the above D) Neither of the above
A
34. For the matched t test, the standard error term is equivalent to: A) the standard error of the mean of the difference scores. B) the standard error of the difference of the means. C) the square root of the pooled variance. D) all of the above
A
39. Which of the following represents the highest achievement performance? A) A percentile rank of 72 on the Graduate Record Examination (given to college graduates) B) A percentile rank of 72 on the Scholastic Aptitude Test (given to high school graduates) C) A percentile rank of 72 on a test of general ability given to all American adults D) None of the above—performance is equal since the percentile ranks are the same
A
4. In a repeated measures ANOVA, as SSsub (the variability of the marginal means of the subjects) increases (and SStotal stays the same): A) SSinter (i.e., SSerror) decreases. B) F decreases. C) the numerator of the F ratio decreases. D) all of the above.
A
4. Which of the following indicates the strongest relationship between two variables? A) -.54 B) .31 C) -.44 D) Cannot be determined without additional information
A
41. The normal curve is defined by: A) a mathematical formula that is never matched exactly by real data. B) a large quantity of raw scores drawn from the same population. C) a large quantity of sample means drawn from the same population. D) a large quantity of standard deviations drawn from the same population.
A
43. A researcher is interested in studying gifted children who have an IQ of 130 or more. As compared to the general population, the variability of the IQ scores for the group of gifted children is: A) small. B) medium. C) large. D) none of the above.
A
47. If we draw one case at random from a population where the raw scores are normally distributed, which of the following is most likely to occur? A) A z score between 0 and -0.25 B) A z score between +1.00 and +1.25 C) A z score between +2.00 and +2.25 D) All of the above are equally likely
A
48. Twenty students take a 10-point quiz, and John obtains a score of 7. At least eight students scored lower than 7. Given only this information, which of the following must be true? A) John's percentile rank cannot be lower than 40. B) John's percentile rank cannot be lower than 55. C) John cannot have received the highest score. D) John's percentile rank must be higher than 50.
A
6. A class takes a 100-point midterm, and the results are: Class Y: 99, 97, 97, 94, 93, 90, 90, 88, 85, 81 If the score of 81 is changed to 31, what will happen to the median? A) It will remain the same. B) It will decrease, but by less than 1 point. C) It will decrease by 5 points. D) Cannot be determined from the information given.
A
6. In two-way analysis of variance, between-cell variance is based on: A) the variability of the cell means. B) the variability of the scores in each cell. C) the total variability of all scores. D) only the variability due to interaction.
A
6. It is more difficult to obtain a highly positive or a highly negative correlation between two variables when: A) the variability on both variables is extremely low. B) the variability on both variables is extremely high. C) the means of the raw scores of the two variables are similar. D) the means of the raw scores of the two variables are very different.
A
10. A set of scores has a mean of 27.4 and a standard deviation of 4.6. If 27.4 is subtracted from each score, what are the new mean and standard deviation? A) Mean = 27.4, standard deviation = 4.6 B) Mean = 0, standard deviation = 4.6 C) Mean = 0, standard deviation = 1 D) Mean = 0, standard deviation = -22.8B
B
11. A researcher wishes to test the hypothesis that the difference between two population means is zero. If the sample means are 106.4 and 99.2, the standard error of the difference is 2.3, and the critical value needed to reject the null hypothesis is 2.09, what should the researcher decide about the difference between the two population means? A) It is equal to 0.0. B) It is not equal to 0.0. C) It is at least 7.2. D) None of the above.
B
12. If a one-way ANOVA is not statistically significant: A) none of the pairwise comparisons will appear to be significant. B) Fisher's protected t tests are not justified. C) LSD is preferred to HSD. D) all of the above.
B
2. In the two-group case, the F ratio for the one-way ANOVA equals: A) the square-root of the corresponding t value. B) the square of the corresponding t value. C) one-half of the corresponding t value. D) twice the corresponding t value.
B
22. What are the practical consequences of a Type II error? A) A false positive—erroneously concluding that the researcher's theory has been supported B) A false negative—erroneously concluding that the researcher's theory has not been supported C) Either a false positive or a false negative, depending on the direction (sign) of the results
B
23. A z score of +2.00 is equal to: A) a T score of 20. B) an SAT score of 700. C) a raw score of 90 on a test with a mean of 75 and a standard deviation of 15. D) all of the above. E) none of the above.
B
23. If the sample size is very small, which of the following is more likely? A) A false positive—erroneously concluding that the researcher's theory has been supported B) A false negative—erroneously concluding that the researcher's theory was not supported C) Both of the above are equally likely D) Neither of the above—sample size has no effect on the likelihood of a false positive or a false negative
B
25. If we are testing a hypothesis about the difference between two population means and we switch from the .05 criterion of significance to the .01 criterion, what happens to the probabilities of a Type I error and a Type II error? A) Both probabilities become smaller. B) The probability of a Type I error becomes smaller and the probability of a Type II error becomes larger. C) The probability of a Type I error becomes larger and the probability of a Type II error becomes smaller. D) Both probabilities become larger.
B
26. If the results of a statistical test indicate that we should retain the null hypothesis, the most reasonable way to state the conclusion is: A) we accept the null hypothesis as true. B) there is not sufficient reason to reject the null hypothesis. C) the experimenter's theory is wrong. D) none of the above are reasonable.
B
3. In one-way analysis of variance, which of the following provides the error term? A) The between-group variance B) The within-group variance C) The total variance D) The difference between the within-group variance and the between-group variance
B
30. The mean of 100 subjects on an attitude scale after receiving a persuasive message is 38.4. The mean for the same subjects before receiving the message was 32.6. If the standard error of the mean of the differences is 1.5 and the critical value needed to reject the null hypothesis is 1.99, the researcher should conclude that: A) the message does not change people's attitudes. B) the message changes people's attitudes. C) there is insufficient evidence to say that the message changes people's attitudes. D) none of the above
B
35. In a 4 × 4 ANOVA with 10 participants in each cell, the total SS is 480. If SSR = 50, SSC = 70, and SSW = 288, how large is the F ratio for the interaction of the two factors? A) 2.25 B) 4.00 C) 8.50 D) 20.0
B
35. Matching subjects in pairs is generally preferable to repeated measures on the same participants, whenever: A) there are simple order effects. B) there are differential carry-over effects. C) the correlation between the two sets of scores is low. D) none of the above.
B
4. In one-way analysis of variance, what can cause participants within the same group to vary? A) The variable whose effects are being studied by the researcher B) Irrelevant factors such as sampling error C) Both of the above D) Neither of the above
B
4. In two-way analysis of variance, what can cause scores within the same cell to vary? A) The independent variables whose effects are being studied by the researcher B) Irrelevant factors such as sampling error C) Both of the above D) Neither of the above
B
4. When drawing inferences about the mean of one population, the 95% confidence interval contains all of the null-hypothesized values of the population mean that should be: A) rejected using the .05 criterion of significance. B) retained using the .05 criterion of significance. C) rejected using the .01 criterion of significance. D) retained using the .01 criterion of significance.
B
44. The entries in the normal curve table in your text tell you: A) The percent area between any two points under the normal curve. B) The percent area between the mean and the scores expressed as z scores. C) The percent area between the mean and the raw scores. D) The percent area between the median and the raw scores.
B
46. In the normal curve, which of the following areas contains the greatest number of cases? A) Between z = -2.00 and z = -1.50 B) Between z = 0 and z = +.50 C) Between z = +1.00 and z = +1.50 D) All of the above contain an equal number of cases
B
49. Ten students take a 10-point quiz, and Sam obtains a score of 8. Seven students scored below 8, one scored 9, and one scored 10. Given only this information, which of the following must be true? A) Sam's percentile rank is 70. B) Sam's percentile rank is 80. C) Sam's percentile rank must be less than 50. D) Sam's percentile rank must be higher than 80.
B
5. If an independent-groups ANOVA and a repeated-measures ANOVA are performed on the same data, which of the following will be the same for both analyses? A) The critical F B) The numerator of the F ratio C) The denominator of the F ratio D) None of the above
B
5. If we compute the correlation between scores on an intelligence test and grades in school, the correlation will be closest to zero for which of the following samples? A) Children chosen randomly from the population B) Gifted children with IQs of at least 130 C) Children from only the fourth through the eighth grades D) Children with IQs between 90 and 130
B
5. In two-way analysis of variance, within-group variance refers to: A) the variability of the cell means. B) the variability of the scores in each cell. C) the total variability of all scores. D) only the variability due to interaction.
B
5. Using a one-way analysis of variance with three groups, a researcher obtains a statistically significant F ratio. If the mean of Group 1 is 16.8, the mean of Group 2 is 10.2, and the value of LSD equals 5.0, what should the researcher conclude about the means of populations 1 and 2? A) They are equal. B) They are not equal. C) They are at least 6.6 points apart. D) There is not sufficient reason to reject the hypothesis that they are equal.
B
6. A set of scores has a mean of 23.8 and a standard deviation of 4.2. If 3 points are subtracted from each score, what are the new mean and standard deviation? A) Mean = 23.8, standard deviation = 4.2 B) Mean = 20.8, standard deviation = 4.2 C) Mean = 23.8, standard deviation = 1.2 D) Mean = 20.8, standard deviation = 1.2
B
7. The 95% confidence interval for the mean of a population is 25.8 to 40.1. Which of the following cannot be the 99% confidence interval for the same population mean? A) 20.6 to 45.3 B) 27.0 to 38.9 C) Both of the above D) Neither of the above
B
8. A set of scores has a mean of 10.5 and a standard deviation of 3.5. If every score is divided by 5, what are the new mean and standard deviation? A) Mean = 2.1, standard deviation = 3.5 B) Mean = 2.1, standard deviation = 0.7 C) Mean = 10.5, standard deviation = 0.7 D) Mean = 10.5, standard deviation = 3.5
B
8. The use of MSw in Fisher's protected t tests is based on the assumption that: A) all of the populations have the same mean. B) all of the populations have the same variance. C) all of the samples have the same size. D) all of the above.
B
8. Which of the following is likely to cause a simple order effect? A) Carry-over from one treatment to the next B) Fatigue C) The use of randomized blocks D) All of the above
B
1. A researcher wishes to test the differences among five population means, using the .05 criterion of significance. Why is it a bad idea to perform 10 t tests for the 10 possible pairs of means? A) The probability of making at least one Type I error will be greater than .05. B) It is too likely that the researcher will get some statistically significant results purely by chance. C) Both of the above. D) Neither of the above.
C
1. Using a one-way analysis of variance with three groups, a researcher obtains a statistically significant F ratio. The group means are: Group 1, 76.3; Group 2, 70.1; Group 3, 74.2. If the value of LSD equals 3.1, the researcher should conclude that which pairs of population means are different? A) Only population 1 and population 2 B) Populations 1 and 2 and populations 1 and 3 C) Populations 1 and 2 and populations 2 and 3 D) Populations 1 and 2, populations 1 and 3, and populations 2 and 3
C
11. A set of scores has a mean of 15.3 and a standard deviation of 3.0. If we first subtract 15.3 from every score and then divide each of the new scores by 3.0, what are the new mean and standard deviation? A) Mean = 5.1, standard deviation = 3.0 B) Mean = 0, standard deviation = 3.0 C) Mean = 0, standard deviation = 1 D) Mean = 5.1, standard deviation = 1
C
11. If a researcher obtains a Pearson r of -.32 and the critical value from a table is .26, what should the researcher decide, and what error might be made? A) Retain the null hypothesis—possible Type I error. B) Retain the null hypothesis— possible Type II error. C) Reject the null hypothesis— possible Type I error. D) Reject the null hypothesis— possible Type II error.
C
11. In a two-tailed test about the mean of one population, the null hypothesis specifies: A) a single value of the sample mean. B) a range of values of the sample mean. C) a single value of the population mean. D) a range of values of the population mean.
C
11. Twenty participants memorize similar lists of words under four different conditions. The total sum of squares is found to be 395. If the SS for subjects equals 59, and the SSerror equals 285, what is the F ratio for the one-way RM ANOVA? A) 0.18 B) 2.82 C) 3.40 D) 3.76
C
12. A researcher who obtains a statistically significant result for the difference between two means should usually conclude that: A) the result has practical importance. B) there is a strong relationship between membership in a population and scores on the variable being studied. C) the result should be supplemented by a measure of effect size. D) the researcher's theory has been contradicted.
C
13. Tukey's HSD test is preferred to Fisher's protected t tests after a one-way ANOVA when: A) all of the samples are the same size. B) there are only three groups. C) there are four or more groups. D) the ANOVA is statistically significant.
C
13. What are the mean and standard deviation of a set of T scores? A) Mean = 0, standard deviation = 1 B) Mean = 10, standard deviation = 1 C) Mean = 50, standard deviation = 10 D) Mean = 500, standard deviation = 100
C
14. Compared to Fisher's protected t tests, Tukey's HSD test: A) leads to a higher rate of Type I errors. B) leads to a lower rate of Type II errors. C) is more conservative. D) maintains less control over experiment-wise alpha.
C
14. The t distribution that you use to find your critical values closely resembles the normal distribution when: A) the sample mean is large. B) the sample variance is large. C) the sample size is large. D) the population standard deviation is large.
C
15. If a mixed-design experiment contains four levels of the between-groups factor and six levels of the repeated-measures factor, and there are 15 participants in each group, how many degrees of freedom are associated with the error term for the repeated-measures factor? A) 210 B) 224 C) 280 D) 300
C
15. If two variables are measured on an entire population, and you are told that Pearson's r for the two variables is near zero, which of the following can you conclude? A) There is very little relationship between the two variables. B) It is unlikely that there is a causal relationship between the two variables. C) The two variables are not linearly related. D) All of the above.
C
15. The critical t value for a one-group experiment will never be: A) less than zero. B) less than 2. C) smaller than the corresponding critical z. D) larger than the corresponding critical z.
C
17. Find Pearson's r, given the following summary statistics: N = 10, ΣXY = 2780, µx = 3.2, µy = 87.5, σx = .4, σy = 7.2. A) +.31 B) -.31 C) -.69 D) -.77
C
17. In one-way analysis of variance, the F ratio is computed by: A) dividing the sum of squares between groups by the sum of squares within groups. B) dividing the sum of squares within groups by the sum of squares between groups. C) dividing the mean square between groups by the mean square within groups. D) dividing the mean square within groups by the mean square between groups.
C
18. For which of the following sets of scores would a grouped frequency distribution be most appropriate? A) Set A: 79, 72, 81, 77, 84, 80, 83, 78, 77, 81 B) Set B: 1,083, 1,279, 406 C) Set C: 97, 46, 82, 65, 77, 89, 67, 58, 92, 87, 75, 74 73, 81, 87, 82, 75, 76, 84, 87, 92, 54, 62, 51, 88, 75, 73, 78, 80 D) Set D: 7, 10, 8, 5, 6, 12, 0, 4, 11, 7, 3, 5, 8, 6, 7, 2, 9, 8, 11, 10, 1, 11, 12, 7, 5, 3, 4, 6, 11, 10, 8, 7, 6, 5, 5, 3
C
18. If you have continuous scores on two variables, converting the scores to ranks and computing the Spearman rank-order correlation coefficient: A) is generally a bad idea because ranks convey less information. B) may produce a better measure of correlation if one variable is highly skewed. C) both of the above. D) neither of the above.
C
18. In a two-way analysis of variance, the denominator of every F test is: A) the sum of squares within cells. B) the mean square between cells. C) the mean square within cells. D) not the same
C
20. If dfw for a 2 × 3 ANOVA is 90, how many participants were in each cell? A) 5 B) 15 C) 16 D) 17
C
21. The 99% confidence interval for the difference between two means is -2.7 to 11.4. Which of the following cannot be the 95% confidence interval for the difference between the same two means? A) -3.2 to 12.1 B) -7.6 to 18.9 C) Both of the above D) Neither of the above
C
21. When the lines on a graph of cell means are not perfectly parallel, you know that: A) the interaction is disordinal. B) there must be some interaction among the population means. C) the F ratio for the interaction will not be zero. D) the F ratio for the interaction will be greater than 1.0.
C
24. A z score of -1.50 is equal to: A) a T score of 30. B) an SAT score of 400. C) a raw score of 67 on a test with a mean of 76 and a standard deviation of 6. D) all of the above. E) none of the above.
C
25. A z score of 0 is equal to: A) a T score of 0. B) a score of 0 on a quiz with a mean of 5.3 and a standard deviation of 10. C) a score of 72 on a test with a mean of 72 and a standard deviation of 6.4. D) all of the above. E) none of the above.
C
25. If you have four equal-sized groups in your study, and the standard deviation for each group is as shown below, how large is MSw? Group s 1 4.9 2 3.6 3 4.1 4 5.4 A) 4.5 B) 6.0 C) 20.74 D) 27.65
C
26. If a student obtains a negative z score: A) the corresponding T score must be negative. B) the corresponding SAT score must be below 400. C) the corresponding raw score must be below the mean. D) all of the above. E) none of the above.
C
27. A researcher wishes to test the hypothesis that a persuasive message about our government will change people's attitudes toward politicians. Each of 20 participants' attitudes is measured before and after receiving the message. The correct procedure to use is: A) the t test with independent samples. B) the z test with independent samples. C) the t test with matched samples. D) the z test with matched samples.
C
29. The mean of 16 subjects on an attitude scale after receiving a persuasive message is 38.4. The mean for the same subjects before receiving the message was 32.6. If the appropriate standard error of the difference is 9.7 and the critical value needed to reject the null hypothesis is 2.145, the researcher should conclude that: A) the message makes no change in people's attitudes at all. B) the message changes people's attitudes. C) there is insufficient evidence to say that the message changes people's attitudes. D) none of the above.
C
3. ALL of the following are true about both one-way analysis of variance and two-way analysis of variance EXCEPT: A) the total variance is divided into between-group variation and within-group variation. B) within-groups variation is considered error variation. C) between-groups variation is divided into three distinct parts. D) the correct statistical model to use is the F distributions.
C
3. If the one-way RM ANOVA is looked at as a two-way ANOVA with subjects as one of the factors, which of the following two-way ANOVA components will equal zero? A) SStotal B) SSbetween-cells C) SSwithin-cells D) SSinter
C
3. Using a one-way analysis of variance with three groups, a researcher obtains a statistically significant F ratio. For which of the following confidence intervals, obtained from the protected t test, should the researcher conclude that the corresponding population means are not equal? A) Group 1 versus group 2: -6.7 to 8.89 B) Group 1 versus group 3: -5.3 to 1.02 C) Group 2 versus group 3: 3.1 to 8.01 D) All of the above E) None of the above
C
3. Which of the following represents the highest score? A) An SAT score of 580 B) A T score of 61 C) A z score of +1.47 D) They are all equal
C
37. Which of the following sets of scores has the largest amount of variability? A) 102.7, 103.2, 104.6, 103.5, 100.9, 101.4 B) 11, 12, 12, 12, 12, 12, 12, 13 C) 1, 0, 6, 9, 12, 14, 3, 8, 17, 2, 5, 6, 15 D) 1,000,000; 1,000,001; 1,000,002; 1,000,004
C
4. A class takes a 10-point quiz, and the results are: Class X: 9, 8, 8, 7, 4, 4, 3, 2, 0. If the score of 9 is changed to 45, what will happen to the mean? A) It will remain the same. B) It will increase by less than 1 point. C) It will increase by 4 points. D) Cannot be determined from the information given.
C
4. If a sample of 25 people has a mean of 5.0 and a standard deviation of 2.5, the estimated standard error of the mean is: A) .1 B) .2 C) .5 D) 1.0
C
4. Using a one-way analysis of variance with three groups, a researcher obtains a statistically significant F ratio. For which of the following confidence intervals, obtained from the protected t test, should the researcher conclude that the corresponding population means are at least 4 points apart? A) Group 1 versus group 2: -7.2 to 5.8 B) Group 1 versus group 3: 0.6 to 9.4 C) Group 2 versus group 3: 4.7 to 8.8 D) Both B and C
C
42. When drawing inferences about the raw scores in a population, the total area under the normal curve: A) increases as the size of the sample increases. B) is equal to plus infinity. C) is defined as 100%. D) has a numerical value equal to the number of cases in the population.
C
43. Why is the normal curve table given in terms of z scores rather than raw scores? A) z scores are more accurate than raw scores. B) All statistical tables are given in terms of z scores. C) Too many tables would be needed if raw scores were used. D) To emphasize that we are using the normal curve model.
C
45. In the normal curve, the area between the mean and a z score of a +1.00 is _____________ the area between a z score of +1.00 and a z score of +2.00. A) smaller than B) equal to C) greater than D) cannot be determined from the information given
C
5. A class takes a 100-point midterm, and the results are: Class Y: 99, 97, 97, 94, 93, 90, 90, 88, 85, 81 If the score of 81 is changed to 31, what will happen to the mean? A) It will remain the same. B) It will decrease, but by less than 1 point. C) It will decrease by 5 points. D) Cannot be determined from the information given.
C
5. If a sample of 100 people has a mean of 34 and a standard deviation of 7, the estimated standard error of the mean is: A) .07 B) .34 C) .7 D) 3.4
C
5. In one-way analysis of variance, what can cause the group means to vary? A) The variable whose effects are being studied by the researcher B) Irrelevant factors such as sampling error C) Both of the above D) Neither of the above
C
52. One hundred students take a 50-point test on which Anne scores 40. If 11 other students also score 40, 68 students score below 40, and the rest score above 40, what is Anne's percentile rank? A) 40 B) 68 C) 80 D) 90
C
6. If the F ratio for the repeated-measures ANOVA turns out to be smaller than the F ratio for an independent-groups ANOVA calculated on the same data, this means that: A) a calculation error has been made. B) the null hypothesis cannot be rejected. C) the RM ANOVA should probably not be performed. D) the subject by treatment interaction is even smaller than can be expected from chance factors.
C
6. The 95% confidence interval for the mean of a population is 87.5 to 113.4. Using hypothesis testing and the .05 criterion of significance, which of the following null hypotheses should be rejected? A) Population mean = 90.2 B) Population mean = 105.7 C) Population mean = 115.6 D) All of the above E) None of the above
C
8. A researcher finds that a correlation of -.51 is statistically significant using the .05 criterion. The researcher should conclude that the population correlation coefficient is: A) equal to zero. B) greater than zero. C) less than zero. D) equal to -.51.
C
8. The 99% confidence interval for the mean of a population is 102.7 to 120.4. Which of the following cannot be the 95% confidence interval for the same population mean? A) 100.3 to 122.8 B) 98.6 to 124.5 C) Both of the above D) Neither of the above
C
9. Differential carry-over effects can be eliminated by: A) counterbalancing. B) using a Latin-Square design. C) matching subjects in blocks. D) using only two repeated conditions.
C
9. Tukey's HSD test assumes that: A) all of the samples have the same mean. B) all of the samples have the same variance. C) all of the samples have the same size. D) all of the above.
C
12. Twenty participants memorize similar lists of words under four different conditions. The total sum of squares is found to be 395, the SS for subjects equals 59, and the SSerror equals 285. What would be the F ratio if the one-way independent-groups ANOVA were calculated, in spite of the fact that this is a repeated-measures design? A) 0.18 B) 2.82 C) 3.40 D) 3.76
D
17. The normal distribution can be used to test hypotheses about population proportions accurately, even with fairly small samples, if: A) the sample proportion is not greater than 1.0. B) the sample proportion is greater than 0. C) the sample proportion is close to 0 or 1.0. D) the sample proportion is close to .5.
D
18. Using which of the following criteria of significance is most likely to lead to a Type II error? A) .10 B) .05 C) .01 D) .001
D
19. A researcher computes a Spearman rank-order correlation coefficient of -.26. The critical value needed to reject the null hypothesis is .38. The correct way to state the conclusion is: A) the population correlation coefficient is equal to zero. B) the population correlation coefficient is not equal to zero. C) the population correlation coefficient is less than zero. D) there is not sufficient reason to believe that the population correlation coefficient differs from zero.
D
2. ALL of the following are transformed scores EXCEPT: A) an SAT score of 530. B) a T score of 47. C) a z score of 0.75. D) a score of 7 on a 10-point quiz.
D
20. A researcher asks two judges to rank 20 patients according to their degree of psychopathology. The researcher would like the judges to be in substantial agreement. Which Spearman rank-order correlation coefficient for the judges' ratings would the researcher most prefer? A) -.89 B) 0 C) .50 D) .63
D
20. In order to perform a one-way analysis of variance, which of the following must be true? A) The sizes of all of the samples must be equal. B) The variances of all of the samples must be equal. C) The means of all of the populations must be equal. D) None of the above.
D
21. A researcher conducts a one-way analysis of variance using four groups and finds that the F ratio is statistically significant. The researcher can conclude that: A) the mean of each population is different from the means of the other three populations. B) the variances of all four populations are the same. C) there is at least one pair of population means that are not the same. D) the variances of all four populations are different.
D
22. If dfw = 50 and dfbet = 4, how many subjects were in each sample? A) 4 B) 5 C) 10 D) 11
D
22. When testing hypotheses about the difference between the means of two populations, the degrees of freedom are: A) N1 + N2 B) N1 - N2 C) N1 + N2 - 1 D) N1 + N2 - 2
D
23. If we retain the null hypothesis about the difference between two population means using the .05 criterion of significance, what is the probability that we have made a Type II error? A) 0 B) .05 C) .95 D) Cannot be determined from the information given
D
24. If your study consists of four groups, each with 15 subjects, and SSbet = 42 and SSw = 35, what does your F ratio equal? A) 4.2 B) 5.6 C) 16.8 D) 22.4
D
24. The power of a statistical test is equal to the probability of: A) retaining the null hypothesis when it is true. B) rejecting the null hypothesis when it is true. C) retaining the null hypothesis when it is false. D) rejecting the null hypothesis when it is false.
D
25. A statistical test that has low power: A) occurs when the sample size is very large. B) is less likely to lead to a Type I error. C) is less likely to lead to a Type II error. D) is likely to be a waste of time and effort.
D
29. The main advantage of a one-tailed test, compared to a two-tailed test, is that: A) only half the calculation is required. B) only half of the calculated t value is required. C) there is only half the risk of a Type I error. D) a smaller critical value must be exceeded.
D
29. When testing a Kruskal-Wallis H for statistical significance, the critical value can be found in the: A) normal curve table. B) t table. C) F table. D) chi square table.
D
31. If a matched t test is carried out and the 95% confidence interval is computed, for which of the following intervals should the researcher retain the null hypothesis of zero change? A) 1.06 to 7.79 B) 10.23 to 16.48 C) Both of the above D) Neither of the above
D
33. For the matched t test, the degrees of freedom are: A) the number of scores. B) the number of pairs. C) the number of scores minus 1. D) the number of pairs minus 1.
D
5. The 95% confidence interval for the mean of a population is 23.4 to 38.6. Using hypothesis testing and the .05 criterion of significance, which of the following null hypotheses should be retained? A) Population mean = 25.7 B) Population mean = 34.2 C) Population mean = 38.1 D) All of the above E) None of the above
D
5. Which of the following represents the lowest score? A) An SAT score of 360 B) A T score of 36 C) A z score of -1.40 D) They are equal
D
6. Using a one-way analysis of variance with three groups, a researcher obtains a statistically significant F ratio. If the mean of Group 1 is 25.7, the mean of Group 2 is 20.4, and the value of LSD equals 8.0, what should the researcher conclude about the means of populations 1 and 2? A) They are equal. B) They are not equal. C) They are 8 points apart. D) There is not sufficient reason to reject the hypothesis that they are equal.
D
7. A researcher finds that a correlation of .32 is not statistically significant using the .05 criterion. The researcher should conclude that the population correlation coefficient is: A) equal to zero. B) likely to be small, but positive. C) equal to .32. D) none of the above.
D
7. A set of scores has a mean of 7.0 and a standard deviation of 3.2. If every score is multiplied by 2, what are the new mean and standard deviation? A) Mean = 7.0, standard deviation = 3.2 B) Mean = 14.0, standard deviation = 3.2 C) Mean = 7.0, standard deviation = 6.4 D) Mean = 14.0, standard deviation = 6.4
D
7. All of the following are advantages of using the mean EXCEPT: A) it takes all of the scores into account. B) it is the most consistent measure of central tendency across different samples drawn from the same population. C) it can be used in many kinds of statistical analyses. D) it is not affected by extreme scores at the high or low end of the distribution.
D
7. Counterbalancing is most likely to be used with which of the following experimental designs? A) Randomized blocks B) A correlational design C) Simultaneous repeated measures D) Successive repeated measures
D
7. Using a one-way ANOVA with three groups, a researcher does not obtain a statistically significant F ratio. If the mean of Group 1 is 20.8, the mean of Group 2 is 23.4, and LSD = 2.4, the correct conclusion to draw with respect to the means of populations 1 and 2 is: A) they are equal. B) they are not equal. C) they are no more than 2.4 points apart. D) there is not sufficient reason to reject the hypothesis that they are equal.
D
73. If the same constant is subtracted from all of the scores in a distribution: A) the standard deviation will be negative. B) that constant will be subtracted from the standard deviation. C) that constant will be added to the standard deviation. D) the standard deviation will be unchanged.
D
8. The median is the best measure of central tendency to use when: A) you plan to use this measure in advanced statistical analyses. B) you want to take all of the scores into account. C) the sample size is very large. D) the exact size of the largest score is unknown.
D
9. If the standard deviation of a set of scores is 4.0 and each score is multiplied by 3, what is the new variance? A) 4.0 B) 12.0 C) 27.0 D) 144.0
D
9. What are the smallest and largest possible values of the standard error of the mean? A) Minus infinity and plus infinity B) -1.00 and +1.00 C) 0 and plus infinity D) 0 and the standard deviation of the population of raw scores
D
28. A sample of N = 100 has a mean of 34.7. The population standard deviation is known to be 6.0. Which of the following null-hypothesized values of the population mean should be rejected using the .05 criterion of significance? A) 33.8 B) 35.0 C) 35.5 D) All of the above E) None of the above
E
16. The number of accidents at work during one month, recorded from all construction companies in the U.S., is likely to produce a distribution that is: A) skewed to the left. B) rectangular. C) unimodal and symmetric. D) a J-curve.
a j-curve
5. As used in statistics, the term population refers to: A) only human beings, such as all American citizens. B) all of the numerical data collected by the researcher. C) a large group of people, cases, or responses that are alike in at least one respect. D) the group of people studied by the researcher.
a large group of people, cases, or responses that are alike in at least one respect
39. Even if your data consist of numbers, it is not valid to perform arithmetic operations on those numbers if your data were derived from: A) a nominal scale. B) a discrete scale. C) a ratio scale. D) an interval scale.
a nominal scale
44. The median is equal to: A) the second quartile. B) the fifth decile. C) the 50th percentile. D) all of the above. E) none of the above.
all of the above
3. The sum of the distances (deviations) of all scores from the mean is: A) always equal to zero. B) always equal to 1.0. C) always equal to N. D) different for different groups of data.
always equal to zero
36. Subjects are randomly assigned to watch either a horror movie or a comedy, and the amount of popcorn they eat during the movie is measured. For this experiment, the type of movie is: A) an independent variable. B) a dependent variable. C) a continuous variable. D) a constant.
an independent variable
7. The standard error of the mean tells us: A) the variability of means of samples of the same size drawn from the same population. B) how trustworthy a single mean is as an estimate of the corresponding population mean. C) both of the above. D) neither of the above
both
9. A frequency distribution is unimodal if it: A) has only one peak. B) has at least two peaks. C) can be divided into two mirror-image halves. D) has a long "tail" at one end.
has only one peak
3. Transformed scores such as z, T, and SAT scores are more likely to give misleading results if the distribution of scores is: A) unimodal and bell-shaped. B) unimodal and symmetric. C) bimodal and symmetric. D) highly skewed.
highly skewed
36. The standard deviation and variance are measures of: A) the general location of a set of scores. B) how spread out the scores in a set of data are from one another. C) both of the above. D) neither of the above.
how spread out the scores in a set of data are from one another
48. Why can't the sum of the deviations from the mean for all of the scores be used as a measure of variability? A) It is in the wrong unit of measurement. B) It will overestimate the variability of the set of scores. C) Deviations from the mean have no effect on variability. D) It is always equal to zero.
it is always equal to zero
41. If you score below average on a midterm exam, you're more likely to get a higher letter grade if the standard deviation is: A) small. B) medium. C) large. D) cannot be determined from the information given.
large
42. A researcher who is interested in introversion-extroversion wants to study the entire range of behavior from extremely introverted to extremely extraverted. The researcher needs a group of subjects for whom the variability of introversion-extraversion is: A) small. B) medium. C) large. D) none of the above.
large
3. One serious problem with grouped frequency distributions is that they: A) require a great deal of time and effort. B) do not look as neat and concise as regular frequency distributions. C) are not suitable for very large samples. D) lose information.
lose information
49. If the variability of a set of scores is small: A) many of the scores will have large deviations from the mean. B) most of the scores will be close to the mean. C) the sum of deviations from the mean will be large. D) the sum of deviations from the mean will be small.
most of the scores will be close to the mean
1. The purpose of a frequency distribution is: A) only descriptive—to conveniently summarize and describe large quantities of data. B) only inferential—to draw conclusions about a group that is too large to measure. C) either descriptive or inferential, depending on the goals of the researcher. D) neither descriptive nor inferential.
only descriptive- to conveniently summarize and describe large quantities of data
7. A regular frequency polygon displays data by using: A) bars. B) cumulative frequencies. C) points on a graph that are connected. D) "stems" and "leaves."
points on a graph that are connected
53. If we divide the sum of squares by the number of scores and then take the square root, we obtain the: A) unbiased variance. B) biased variance. C) population variance. D) population standard deviation.
population standard deviation
13. A very difficult exam would be likely to produce a distribution that is: A) negatively skewed. B) positively skewed. C) unimodal and symmetric. D) rectangular.
positively skewed
17. If a roulette wheel is fair, the number of times each number pays off would produce a distribution that is: A) skewed to the left. B) rectangular. C) unimodal and symmetric. D) a J-curve.
rectangular
13. A Type I error occurs when the researcher: A) rejects a false null hypothesis. B) rejects a true null hypothesis. C) retains a false null hypothesis. D) retains a true null hypothesis.
rejects a true null hypothesis
16. If we convert a set of raw scores to z scores, the shape of the distribution will: A) become more symmetric. B) become more skewed. C) become more like the bell-shaped normal curve. D) remain the same.
remain the same
4. Why are inferential statistics used in the behavioral sciences? A) To correct for the error in tests, questionnaires, and other research instruments. B) A science must use numerical values. C) Researchers need a way to summarize the large amounts of data that they obtain. D) Researchers want to draw conclusions about groups that are too large to measure in their entirety.
researches want to draw conclusions about groups that are too large to measure in their entirety
14. A Type II error occurs when the researcher: A) rejects a false null hypothesis. B) rejects a true null hypothesis. C) retains a false null hypothesis. D) retains a true null hypothesis.
retains a false null hypothesis
55. When computing the standard deviation, the last step is to take the square root. This is done to: A) prevent the result from being negative. B) return to the original unit of measurement. C) provide a more accurate estimate of the population. D) all of the above.
return to the original unit of measurement
40. If you score above average on a midterm exam, you're more likely to get a higher letter grade if the standard deviation is: A) small. B) medium. C) large. D) cannot be determined from the information given.
small
2. If we draw 1,000 samples of size 100 from a population and compute the mean of each sample, the variability of the distribution of sample means will tend to be _________ the variability of the raw scores in any one sample. A) smaller than B) equal to C) greater than D) cannot be determined from the information given
smaller than
50. If we subtract the mean from each score, square each of the differences, and then add them, we obtain the: A) standard deviation. B) variance. C) sum of squares. D) range.
sum of squares
1. The purpose of descriptive statistics is to: A) draw conclusions about a group that is too large to measure. B) summarize important characteristics of a set of data. C) both of the above. D) neither of the above.
summarize important characteristics of a set of data
45. The range is defined as: A) the average distance between each pair of scores. B) the average distance of each score from the mean. C) the largest distance of a score from the mean. D) the largest score minus the smallest score.
the largest score minus the smallest one
30. In a symmetric distribution: A) the mean is smaller than the median. B) the mean is larger than the median. C) the mean and the median are equal. D) none of the above.
the mean and the median are equal
1. SAT scores, T scores, and z scores build what information into the score itself? A) The mean and standard deviation B) The median and standard deviation C) Only the mean D) Only the standard deviation
the mean and the standard deviation
34. It is not possible for more than half of the scores in a distribution to be above: A) the mode. B) the median. C) the mean. D) the mode, median, or mean.
the median
46. Which of the following represents the best performance? A) The first quartile B) The fourth decile C) The median D) The 30th percentile
the median
31. Which of the following is almost never used in the behavioral sciences because it is often a poor measure of central tendency? A) The mean B) The median C) The mode D) None of the above (i.e., all are used often)
the mode
10. In hypothesis testing, which of the following is assumed to be true in order to find the p value? A) The alternative hypothesis B) The null hypothesis C) Both of the above D) None of the above
the null hypothesis
44. Which of the following is almost never used in the behavioral sciences because it is often a poor measure of variability? A) The standard deviation B) The variance C) The range D) None of the above
the range
6. If we draw two samples from the same population, one with 100 cases and one with 25 cases, for which sample will the estimate of the standard error of the mean tend to be smaller? A) The sample with 25 cases B) The sample with 100 cases C) The standard error of the mean will be the same for both samples because they come from the same population D) Cannot be determined from the information given
the sample with 100 cases
39. Which of the following cannot be negative? A) The mean B) The median C) The mode D) The standard deviation
the standard deviation
45. Which of the following represents the best performance? A) The 40th percentile B) The first decile C) The median D) The third quartile
third quartile
15. Scores on a standardized intelligence test are likely to produce a distribution that is: A) skewed to the left. B) skewed to the right. C) unimodal and symmetric. D) a J-curve.
unimodal and symmetric
51. If we subtract the mean from each score, square each of the differences, add them, and then divide this sum by the number of scores, we obtain the: A) standard deviation. B) variance. C) sum of squares. D) range.
variance
34. Why is it useful to convert a raw score to a percentile rank? A) We can draw inferences about a population that is too large to measure. B) We can summarize a group of scores more concisely. C) We can show how the raw score compares to the scores of a specific group. D) We can show the percent of total points that was obtained.
we can show how the raw score compares to the scores of a specific group
38. If all of the variables in your study have been measured on a nominal scale: A) your study is not a true experiment. B) you can use only descriptive statistics. C) you can use only parametric statistics. D) you can use only nonparametric statistics.
you can use only nonparametric statistics
2. If we were able to measure the entire population, the standard error of the difference would be: A) zero. B) 1.0. C) equal to the pooled variance. D) equal to the standard deviation of the raw scores.
zero
75. Before conducting a maze-learning experiment, you weigh two groups of rats. The experimental group, which contains 20 rats, has an average weight of 18 ounces, while the control group, which contains 10 rats, has an average weight of 15 ounces. What would the average weight be if the two groups of rats were combined into one large group? A) 16 B) 16.5 C) 17 D) 17.5
17
4. When constructing a grouped frequency distribution, the total number of intervals should be approximately: A) 2 to 5. B) 5 to 10. C) 8 to 15. D) 5 to 25.
8 to 15
19. On a midterm exam, your score expressed as a T score is 62. The mean of the exam is 72, and the standard deviation is 5. Your raw score is: A) 66. B) 67. C) 77. D) 78.
78
14. A researcher finds that the Pearson r between two variables is +1.17. This indicates that: A) an error in computation was made. B) there is a strong positive correlation between the two variables. C) one of the variables is not normally distributed. D) both of the variables are not normally distributed.
A
14. An increase in subject-to-subject variability in a mixed design (i.e., MSW) will cause a reduction in the F ratio associated with: A) the main effect of the between-groups factor. B) the main effect of the within-subjects factor. C) the interaction of the two factors. D) both B and C.
A
72. If all of the scores in a distribution are multiplied by the same constant: A) the standard deviation will be multiplied by that constant. B) the standard deviation will be divided by that constant. C) the standard deviation will be equal to that constant. D) the standard deviation will be unchanged.
A
9. Using a sample of 10,000 cases, a researcher obtains a Pearson r of .11 that is statistically significant using the .05 criterion. The researcher should conclude that: A) the population correlation coefficient is not zero. B) there is a strong positive relationship between the two variables. C) the relationship between the two variables is very close to being linear. D) all of the above.
A
9. When estimating the standard error of the difference, the pooled variance: A) gives greater weight to the larger of the two samples. B) gives greater weight to the sample with the larger variance. C) gets smaller as the sample size gets larger. D) all of the above.
A
26. When using the two-group t test with large samples, the results will not be accurate if: A) the dependent variable is not normally distributed within each population. B) the two sample variances differ substantially, and the sample sizes are far from being equal. C) both of the above. D) neither of the above.
B
27. A researcher using the Kruskal-Wallis H test hopes to obtain statistical significance. The researcher would like the sums of the ranks of the groups in the study to be: A) equal to each other. B) very different from each other. C) close to the expected sums of the ranks. D) equal to N (N + 1)/2.
B
27. If we transform a set of raw scores to z scores and plot the z scores as a regular frequency polygon, the shape of the distribution will be the normal curve: A) always—graphing a set of z scores always produces the normal curve. B) only if the distribution of raw sores was a normal curve. C) never—a set of z scores cannot take the form of the normal curve.
B
3. If two variables are highly positively correlated, which of the following must have approximately the same numerical value and the same sign? A) The paired raw scores B) The paired z scores C) Both of the above D) Neither of the above
B
12. If a researcher obtains a Pearson r of .21 and the critical value from a table is .26, what should the researcher decide, and what error might be made? A) Retain the null hypothesis— possible Type I error. B) Retain the null hypothesis— possible Type II error. C) Reject the null hypothesis— possible Type I error. D) Reject the null hypothesis— possible Type II error.
B
16. For which of the following should the protected rank-sum test be used as a follow-up test when the overall results are statistically significant? A) The rank-sum test B) The Kruskal-Wallis H test C) The one-way ANOVA D) All of the above
B
18. To create a confidence interval for the proportion in a population, you need to use your sample proportion to estimate: A) the level of confidence that you need. B) the standard error of the proportion. C) the critical values. D) the number of degrees of freedom.
B
19. A researcher conducts a two-way analysis of variance. Factor 1 is caffeine dosage (high or low), factor 2 is gender (male or female), and the dependent variable is scores on a mathematics test. Which of the following indicates an interaction effect? A) Men who receive a high dose of caffeine and women who receive a high dose of caffeine score higher on the test. B) Men who receive a high dose of caffeine and women who receive a low dose of caffeine score higher on the test. C) Both of the above. D) Neither of the above.
B
19. In a one-way analysis of variance, if the null hypothesis is true, what value of F should we expect in our observed data (on average)? A) 0 B) 1.0 C) N (the total number of scores) D) N - 1
B
19. Why do researchers usually not use the .001 criterion of significance? A) It would make a Type I error too likely. B) It would make a Type II error too likely. C) It would make both a Type I and a Type II error too likely. D) None of the above.
B
50. Fifty students take a 100-point midterm exam on which Tricia scores 86. If 40 students obtained scores below 86, what is Tricia's percentile rank? A) 40 B) 80 C) Cannot be determined from the information given but must be at least 80. D) Cannot be determined from the information given but cannot be higher than 80.
C
12. When testing a hypothesis about the mean of one population, the only way to avoid sampling error is to: A) use the standard error of the mean. B) use a random sample. C) draw a large number of samples. D) measure the entire population.
D
13. A researcher theorizes that intelligence is not correlated with introversion-extraversion. The researcher obtains a sample of 15 people and finds that the correlation between intelligence and introversion is .37. This Pearson r is not statistically significant, so the researcher concludes that her hypothesis has been supported. What is wrong with this conclusion? A) No firm conclusion can be drawn because the probability of a Type II error is unknown. B) The sample size is too small, making useful results unlikely. C) A researcher's theory should usually not be identified with the null hypothesis. D) All of the above.
D
13. A t-test is used in place of the z-score for groups when which of the following is not known? A) The sample mean B) The population mean C) The sample standard deviation D) The population standard deviation
D
14. What are the mean and standard deviation of a set of SAT scores? A) Mean = 0, standard deviation = 1 B) Mean = 10, standard deviation = 1 C) Mean = 50, standard deviation = 10 D) Mean = 500, standard deviation = 100
D
15. A frequency distribution contains 40 scores that vary from 124 to 197. If we subtract 100 from each score, which of the following will NOT change? A) The mean B) The median C) The mode D) The standard deviation
D
15. In a one-way analysis of variance, the degrees of freedom for the numerator is equal to: A) the number of scores. B) the number of groups. C) the number of scores minus one. D) the number of groups minus one.
D
16. A researcher finds that the correlation between obesity and satisfaction with life in general is -.74 and that this correlation is statistically significant using the .05 criterion. The researcher can conclude that: A) being obese (i.e., overweight) makes people unhappy with life in general. B) being unhappy with life in general causes people to become overweight. C) both of the above. D) neither of the above.
D
16. In a one-way ANOVA, mean squares are obtained by dividing each sum of squares by: A) the total number of scores. B) the number of different groups. C) the total number of scores minus one. D) its associated degrees of freedom.
D
16. Using the .01 criterion of significance, the probability of making a Type II error is: A) .01. B) .05. C) .99. D) Cannot be determined from the information given.
D
18. The smallest and largest possible values of F are: A) minus infinity and plus infinity. B) -1.0 and +1.0. C) 0 and 1.0. D) 0 and plus infinity.
D
2. If 100 students take a 10-point quiz, the most appropriate way to summarize this information is to use: A) a grouped frequency distribution. B) a regular frequency distribution. C) either a regular or grouped frequency distribution, depending on the goals of the researcher. D) neither a regular nor a grouped frequency distribution.
a regular frequency distribution
2. If we specify a range of values within which a parameter (such as a population mean) is likely to fall, this range can be referred to as: A) an interval estimate. B) a range statistic. C) a point estimate. D) a statistical estimate.
an interval estimate
6. As used in statistics, the term sample refers to: A) a relatively small number of people drawn from a specified population. B) all of the cases about which the researcher wishes to draw conclusions. C) the tests, questionnaires, and other instruments used by the researcher. D) the minimum number of people that a research study should include.
a relatively small number of people drawn from a specified population
36. Which of the following is a raw score? A) An SAT score of 530 B) A percentile rank of 82 C) A z score of +1.10 D) A score of 7 on a 10-point quiz
a score of 7 on a 10-point quiz
12. A distribution that is skewed to the right would have: A) a small number of very low scores. B) a small number of very high scores. C) both of the above. D) neither of the above.
a small number of very high scores
40. If a subject's attitude toward the death penalty is assessed on a scale that consists of: strongly opposed, somewhat opposed, slightly opposed, neutral, slightly for, somewhat for, and strongly for, which of the following types of scales is being used? A) A nominal scale B) An ordinal scale C) An interval scale D) A ratio scale
an ordinal scale
1. As the size of the samples increases, the standard error of the difference: A) becomes smaller. B) remains the same. C) becomes larger. D) cannot be determined from the information given.
becomes smaller
8. A frequency distribution is symmetric if it: A) has only one peak. B) has at least two peaks. C) can be divided into two mirror-image halves. D) has a long "tail" at one end.
can be divided into two mirror-image halves
3. You need a convenient way to tell a friend about all of the grades that you have received in college, so you say that your grade-point average is 3.27. This is an example of a(n) A) descriptive statistic. B) inferential statistic. C) parameter. D) population statistic.
descriptive statistic
2. The purpose of inferential statistics is to: A) draw conclusions about a group that is too large to measure. B) summarize important characteristics of a set of data. C) draw inferences about the sample that the researcher is studying. D) none of the above.
draw conclusions about a group that is too large to measure
29. In a positively skewed distribution: A) the mean is smaller than the median. B) the mean is larger than the median. C) the mean and the median are equal. D) none of the above.t
he mean is larger than the median
33. If all of the scores in a distribution are multiplied by the same constant: A) the mean will be multiplied by that constant. B) the mean will be divided by that constant. C) the mean will be equal to that constant. D) the mean will be unchanged.t
he mean will be multiple by that constant
11. Which of the following describes a J-curve? A) Everyone receives exactly the same score. B) The distribution has two peaks, one much larger than the other. C) Most people score zero, some people score one, a few people score two, and very few people score higher than two. D) Most people score at the center of the distribution, and as you go farther from the center, there are fewer scores.
most people score zero, some people score one, a few people score two, and very few people score higher than two
10. A researcher wishes to test the hypothesis that the difference between two population means is zero. If the sample means are 106.4 and 99.2, the standard error of the difference is 6.1, and the critical value needed to reject the null hypothesis is 2.09, what should the researcher decide about the difference between the two population means? A) It is equal to 0.0. B) It is not equal to 0.0. C) It is at least 7.2. D) None of the above.
none of the above
6. A bar chart would be more appropriate than a histogram for displaying: A) scores on an IQ test. B) family income. C) results of a 100-point midterm exam. D) number of children per family.
number of children per family
37. Psychological variables, such as anxiety or creativity, are not likely to be measured: A) on an ordinal scale. B) using numbers. C) on a ratio scale. D) as a continuous variable.
on a ratio scale
7. Researchers in the behavioral sciences usually wish to draw conclusions about: A) descriptive statistics. B) inferential statistics. C) one or more populations. D) one or more samples.
one or more populations
38. The percent of cases in a specific reference group scoring at or below a given score is called a: A) percentile. B) percentile rank. C) transformed score. D) z score.
percentile rank
