Your midterm exam will consist of two parts, a practical SPSS application test which will be worth 40% of the test grade, and a multiple choice part which will be worth 60%. The practical SPSS test will consist of three statistical applications: a t-test, a chi-square analysis, and a univariate analysis of variance, worth ten, twelve, and eighteen points, respectively.

The multiple choice part of the test will consist of 40 questions drawn from this list. They will be worth 1.5 points each. The questions come from the class slides and the two required texts.

  1. Values used to make inferences about the characteristics of the population from which they were drawn, including the variation of the sample characteristics from corresponding population parameters are called
  2. Descriptive statistics
  3. Inferential statistics
  4. Goodness-of-fit measures
  5. Population parameters
  6. A list of all the registered Republican voters in Los Angeles Country is an example of a
  7. Population parameter
  8. Independent variable
  9. Macro-level construct
  10. Sampling frame
  11. Quasi-experimental studies differ from experimental studies in that in the quasi-experimental study the experimenter
  12. Has control over assignment of cases to conditions of the independent variable
  13. Has control over assignment of cases to conditions of the dependent variable
  14. Has control over when and where the dependent variable can be measured
  15. Has no control over either dependent variable or independent variable
  16. Which type of research relies on random assignment?
  17. Experimental
  18. Quasi-experimental
  19. Naturalistic
  20. Correlational
  21. Which of the following could describe “reliability” in the context of quantitative research?
  22. Test-retest correlation
  23. Internal consistency of items on a test
  24. Correlation with other measures of the same thing
  25. A and B above
  26. B and C above
  27. A and C above
  28. None of the above
  29. Which of the following is an example of cluster sampling?
  30. Randomly drawing a name out of a hat
  31. Drawing randomly from randomly drawn lists of neighborhoods or zip codes
  32. Sampling from a list which has proportional representation of ethnic groups or other demographic characteristics in proportion to their numbers in the population
  33. None of the above
  34. Identify the independent variable in the following hypothesis: the impact of gender on job category is thought to be mediated by educational attainment
  35. Gender
  36. Job category
  37. Educational attainment
  38. None of the above
  39. Identify the dependent variable in the following hypothesis: the impact of gender on job category is mediated by educational attainment
  40. Gender
  41. Job category
  42. Educational attainment
  43. None of the above
  44. Identify the control variable in the following hypothesis: the impact of gender on job category is mediated by educational attainment
  45. Gender
  46. Job category
  47. Educational attainment
  48. None of the above
  49. Which of the following is an example of a categorical variable?
  50. Type of communications medium
  51. Political attitudes
  52. Income
  53. None of the above
  54. Which of the following is an example of a continuous variable?
  55. Number of hours spent watching TV every day
  56. Gender
  57. Ethnicity
  58. All of the above
  59. Which of the following is an example of interval level measurement?
  60. IQ test
  61. Gender
  62. Weight
  63. Political Affiliation
  64. Which of the following is an example of ratio level measurement?
  65. Political affiliation
  66. Hours spent surfing the Net
  67. Preference among brands of cereal
  68. All of the above
  69. Which of the following is an example of nominal level measurement?
  70. Ethnicity
  71. Rankings in a beauty contest
  72. Temperature
  73. Income in dollars
  74. A summary which gives an account of how often answers in each category of responses to a question occur within a sample is called a
  75. Dendogram
  76. Normal distribution
  77. Kurtosis
  78. Frequency distribution
  79. In the picture below, how would the summary of the data represented in the picture be described?
  80. A homogeneous distribution
  81. A heterogenous distribution
  82. A normal distribution
  83. None of the above
  1. A distribution in which most of the cases occur at the low end of the scale is called
  2. Negatively skewed
  3. Positively skewed
  4. Leptokurtic
  5. Platykurtic
  6. A distribution in which most of the cases are peaked around the mean is called
  7. Negatively skewed
  8. Positively skewed
  9. Leptokurtic
  10. Platykurtic
  11. Which of the measures below does this describe: Measure which is most stable for random samples, which makes it suitable for making estimates about populations from samples. It has the property that the sum of the deviations of the raw scores from it equals zero.
  12. Standard deviation
  13. Mode
  14. Mean
  15. Median
  16. Which of the measures below does this describe: the response value for which there are an equal number of responses both below and above it (e.g., larger or smaller). Used with ordinal or numerical variables
  17. Standard deviation
  18. Mode
  19. Mean
  20. Median
  21. Which of the measures below does this describe: the most frequently selected (commonly occurring) response category. The only measure for nominal level variables but can be used with scaled data.
  22. Standard deviation
  23. Mode
  24. Mean
  25. Median
  26. What is the effect of an “outlier” on the mean?
  27. Makes it more different from the median
  28. Makes it more similar to the median
  29. Keeps the distribution from being used for inferential statistics
  30. None of the above
  31. What is the property of a sample described here: the degree or amount of variability in a set of responses to a quantitative measure such as a questionnaire item
  32. The significance
  33. The mean
  34. Dispersion
  35. Kurtosis
  36. The index of qualitative variation would be used for what kind of data?
  37. Nominal
  38. Ordinal
  39. Ratio
  40. Interval
  41. What is the following a definition of? the sum of the squared deviations from the sample mean, divided by N-1 where N is the number of cases.
  42. The mean
  43. The harmonic mean
  44. The dispersion
  45. The variance
  46. What is the square root of the property described in 25 called?
  47. The standard deviation
  48. The interquartile point
  49. The range
  50. The variance
  51. The mean, median and mode of responses all coincide in a
  52. Platykurtic distribution
  53. Normal distribution
  54. Skewed distribution
  55. None of the above
  56. What is the measure that has the following properties: it allows you to make comparisons between samples with respect to their variability (how much a respondent from the sample typically departs from the mean). Its size is generally about one-sixth the size of the value of the range
  57. The standard deviation
  58. The interquartile point
  59. The harmonic mean
  60. The variance
  61. What is this a definition of? deviation of a raw score from the mean in standard deviation units
  62. A standard score
  63. A z score
  64. Both A and B
  65. Neither A nor B
  66. In a normal distribution, what percentage of scores fall above the mean?
  67. 68%
  68. 34%
  69. 50%
  70. 95%
  71. About what percent of cases fall within 2 SDs of the mean?
  72. 68%
  73. 34%
  74. 50%
  75. 95%
  76. According to the Central Limit Theorem, the larger the sample size, the greater the probability that the obtained sample mean will ______the population mean
  77. Depart from
  78. Be the opposite of
  79. Be the standard deviation of
  80. Approximate
  81. In the normal table (“Area under the Normal Curve” ) you look up a Z score of 2.2 and to the right of that you find the “area between the mean and Z” is 48.61. Thus 48.61% of the cases in the normal distribution lie between the mean and Z=2.2. What proportion of cases lie below this?
  82. 34.39%
  83. 1.39%
  84. 98.61%
  85. 50. 61%
  86. What proportion of cases lie above this?
  87. 34.39%
  88. 1.39%
  89. 98.61%
  90. 50. 61%
  91. The standard deviation of the sampling distribution of sample means is called the
  92. Standard error of the mean
  93. Variability of the mean
  94. Alpha coefficient
  95. Variance
  96. What is the term for how much statistics can be expected to deviate from parameters when sampling randomly from the population
  97. Standard error of the mean
  98. Variability of the mean
  99. Alpha coefficient
  100. Variance
  101. What is the relationship between sample size and the property described in (36) above?
  102. It gets greater with increased sample size
  103. It gets greater as the square of sample size increases
  104. It gets smaller as the inverse of sample size increases
  105. It gets smaller with increased sample size
  106. Suppose we had the following data: 5,6,7,8, 9. and we calculated their mean as 35/5 = 7. What are the degrees of freedom in computing the mean?
  107. 5
  108. 4
  109. 3
  110. 1
  111. Suppose we wanted to construct a confidence interval around the mean of 2969.56 such that we can have 95% confidence that the population mean for the variable “vehicle weight” will fall within this range. To obtain this confidence interval, we would need to take the mean and add to it the quantity (1.96 times the standard error). What does the number 1.96 represent in this case?
  112. A constant which would apply to all such calculations
  113. The beta weight associated with the variable
  114. The risk area under the normal curve corresponding to 5%
  115. The difference between the mean weight and the next higher weight
  116. A ______variable is one on which each case is coded for either presence or absence of the attribute. For example, we could recode the ethnicity data into the ______variable “whiteness” or “Chinese-ness” so that every case would have either a 1 or a zero on the variable. All of the white (or Chinese) respondents would get a 1 and the others would get a zero on the variable. This kind of variable is called a
  117. Dummy variable
  118. Control variable
  119. Composite variable
  120. Mediator variable

41. Which of the following is an example of a null hypothesis?

a. There is no difference between males and females in attitudes toward voting

b. Males and females differ in their attitudes towards voting

c. Males tend to vote more often than females

d. The relationship between sex and voting is unknown

42. Statistical significance is

a. an indicator of the importance of the relationship between two variables

b. an indicator of the probability of a test statistic being the result of chance alone.

c. a sign that your sample was drawn randomly

d. a sign that you have eliminated random error

43. A student calculated a Chi square statistic with a significance level of .01 for a table relating gender to voting behavior for those 40-50 years old. She calculated a Chi square statistic with a significance level of .25 for the same two variables for those 20-30 years old. Which of the following best summarizes the findings?

a. She can be more sure that the obtained relationship between gender and voting behavior among those 40-50 years old was not just a chance result than the obtained relationship for those 20-30 years old.

b. She can be relatively sure that the relationship between gender and voting behavior is strong

c. The relationship between gender and voting behavior is stronger for those 40-50 years old than for those 20-30 years old

d. The level of association between gender and voting behavior is quite low for both groups.

44. We are conducting an empirical study that examines the relationship of communication frequency and relationship stability in couples married for at least 3 years. Which is the independent variable? Which is the dependent variable?

a. independent variable: relationship stability

dependent variable: communication frequency

b. independent variable: length of marriage

dependent variable: relationship stability

c. independent variable: communication frequency

dependent variable: length of marriage

d. can’t tell from the information given

  1. What kinds of question should we not explore through a contingency table?
  2. What are the differences in job classification attributable to gender?
  3. What is the impact of gender on socio-economic status?
  4. How does temperature affect annual rainfall?
  5. How does job classification affect preference for news source?
  6. In a chi-square analysis the principal interest is in comparing the obtained frequencies to the
  7. Column frequencies
  8. Row marginals
  9. Obtained marginals
  10. Expected frequencies
  11. Consider the contingency table below: What appears to be the relationship between educational attainment and employment category?

  1. Educational attainment has an impact on job category only for elementary school graduates
  2. Educational attainment is associated with employment category.
  3. Only clerical work is affected by educational attainment.
  4. None of the above.
  1. Lambda is a
  2. Measure of association for interval level variables
  3. Reliability coefficient
  4. Measure of proportional reduction of error for contingency tables
  5. Measure only used for ordinal level variables
  6. Lambda is sometimes flawed as a measure because it
  7. Ranges from -1 to +1
  8. Doesn’t test significance
  9. Comes out to zero quite a lot
  10. Is too hard to calculate
  11. An alternative to lambda which SPSS reports is
  12. Alpha
  13. Tau
  14. Chi-square
  15. Gamma
  16. One of the things we can do in a cross-tabulation analysis is to look for the effect of a control variable on the relationship between an independent and dependent variable. In the table below, which is the control variable?

  1. gender
  2. educational attainment
  3. employment category
  4. status as a manager
  1. Which of the following is not a step in statistical hypothesis testing?
  2. Specify the research hypothesis and corresponding null hypothesis
  3. Compute the value of a test statistic about the relationship between the two hypotheses
  4. Calculate the DF and look up the statistic in the appropriate distribution to see if it falls into the critical region
  5. If the result is not significant, move the critical region lower until you reach significance
  6. The null hypothesis with respect to the relationship between two variables is that
  7. The population and the sample means are different
  8. The variables are independent of one another
  9. There is no way to determine if the relationship is significant
  10. The two variables are related to one another
  11. Which of the following statements about chi-square is correct?
  12. Chi-square ranges between -1 and 1
  13. Chi-square can be interpreted as an index of the proportional reduction of error
  14. Chi-square is a measure of the statistical independence of two variables
  15. The larger the value of chi-square for a constant value of DF, the less the dependence of the two variables (the weaker their association)
  16. Which of the following is a way of controlling for the influence of extraneous variables in an experiment?
  17. The case-control method
  18. Randomization
  19. Large numbers of subjects
  20. Test-retest reliability
  21. The ability to show that the causal impact of an independent variable on a dependent variable is legitimate and not attributable to other extraneous and uncontrolled variables is called
  22. Construct validity
  23. Statistical conclusion validity
  24. External validity
  25. Internal validity
  26. A “manipulation check is” a way of ensuring adequate
  27. Construct validity
  28. Statistical conclusion validity
  29. External validity
  30. Internal validity
  31. Features of an experimental setting or questionnaire which induce people to behave in an artificial way are called
  32. Demand characteristics
  33. Debriefing
  34. Experimental attrition
  35. Normative role decay
  36. Which of the following is an example of method variance?
  37. Changing the gender of the experimenter when gender is not a variable in the study
  38. Using paper and pencil measures on one occasion and an interview on another
  39. Using different incentives for different subjects when incentive is not a variable
  40. All of the above
  41. None of the above
  42. Which of the following types of scale does this describe: an object of judgment is evaluated against a set of rating scales (usually five to seven steps) with bi-polar adjectives at either end, such as good-bad or friendly-unfriendly
  43. Likert scale
  44. Guttman scale
  45. Rausch scale
  46. Semantic differential scale
  47. Which of the following types of scales does this describe: people are asked to indicate if they strongly agree, agree, are neutral, disagree, or strongly disagree with declarative statements about a topic.
  48. Likert scale
  49. Rausch scale
  50. Semantic differential scale
  51. None of the above
  52. Which of the following might be a problem with the way that people fill out questionnaires?
  53. They prefer odd numbers to even ones
  54. The have a tendency to disagree with statements rather than agree with them
  55. They give their best responses toward the end of the questionnaire
  56. They may reject items with “always” and “never”
  57. Underlying the t statistic is a sampling distribution of
  58. Sample means
  59. Population means
  60. Differences of sample means
  61. Standard error of sample means
  62. The ttest is for comparing
  63. The difference of means for two independent groups
  64. The difference of means for dependent samples
  65. The difference of a sample mean and a population mean
  66. All of the above
  67. None of the above
  68. What is the DF of a t test comparing two independent groups where one group has a N of 20 and the other an N of 38?
  69. 18
  70. 58
  71. 56
  72. 760
  73. In conducting a t-test, the researcher can do a one-tailed or a two-tailed test. Under what circumstances would a one-tailed test be conducted?
  74. When the researcher hoped to have a bigger critical area for getting significance
  75. When the researcher had predicted the direction of the mean differences
  76. When the sample was not normal with respect to the underlying distribution
  77. None of the above
  78. If I do a two-tailed test of my hypotheses and set the confidence level to .05, what area under the normal curve does my obtained value of t have to fall in to obtain significance?
  79. The upper 5% of either end of the distribution
  80. The upper 2.5% of either end of the distribution
  81. The upper ten percent of either end of the distribution
  82. The upper 5% of whatever end I predict it’s going to fall in
  83. In certain cases, for example in “before and after” designs or when members of group A have been matched with members of group B on all salient characteristics except one, the variable of interest, an alternative formula for computing t is used. What is the main difference of this t from the t for independent samples?
  84. t is based on the departure of the difference scores from the mean difference score
  85. You use a different menu option in SPSS
  86. The scores in post-test groups are known to be higher
  87. All of the above
  88. We find out if the ______in two groups are equal before deciding on what sort of t-test we will perform
  89. Means
  90. Standard errors
  91. Medians
  92. Variances

70. An analysis of variance looks for the causal impact of a nominal level independent variable (factor) on

a. A nominal variable

b. An ordinal variable

c. An interval or ratio level variable