Deck 11: Analysis of Quantitative Data

Question 1

Discuss the concepts of control variables and trivariate tables.What are three limitations of trivariate tables?

Answer 1

● In order to meet all the conditions needed for causality,researchers want to "control for" or see whether an alternative explanation explains away a causal relationship.If an alternative explanation explains a relationship,then the bivariate relationship is spurious.Alternative explanations are operationalized as third variables,which are called control variables because they control for alternative explanations.
● A trivariate table has a bivariate table of the independent and dependent variable for each category of the control variable.These new tables are called partials.The number of partials depends on the number of categories in the control variable.Partial tables look like bivariate tables,but they use a subset of the cases.Only cases with a specific value on the control variable are in the partial.Thus,it is possible to break apart a bivariate table to form partials,or combine the partials to restore the initial bivariate table.
● Trivariate tables have three limitations.First,they are difficult to interpret if a control variable has numerous categories.Second,control variables can be at any level of measurement,but interval or ratio control variables must be grouped (i.e.,converted to an ordinal level),and how cases are grouped can affect the interpretation of effects.Finally,the total number of cases is a limiting factor because the cases are divided among cells in partials.

Question 2

Is a Type I or Type II error more likely if a 0.05 level is used? Explain.

Answer 2

● Type I error: Falsely accepting the null hypothesis when in fact there is a causal relationship (usually occurs at a more precise level such as at the 0.01 level)
● Type II error: Indicates a relationship when in fact no causal relationship exists (random factors actually caused the results and usually occurs at the 0.10 level)
● 0.05 level is a compromise between Type I and Type II errors.

Question 3

Describe each of the three measures of central tendency.What are the main differences between them? How are they affected by a normal versus a skewed distribution of data?

Answer 3

● Mean: A measure of central tendency for one variable that indicates the arithmetic average (i.e.,the sum of all scores divided by the total number of scores)
● Median: A measure of central tendency for one variable indicating the point or score at which half the cases are higher and half are lower
● Mode: A measure of central tendency for one variable that indicates the most frequent or common score
● Skewed distribution: If the frequency distribution forms a "normal" or bell-shaped curve (normal distribution),the three measures of central tendency equal each other.If the distribution is a skewed distribution (i.e.,more cases are in the upper or lower scores),then the three will not be equal.If most cases have lower scores with a few extreme high scores,the mean will be the highest,the median in the middle,and the mode the lowest.If most cases have higher scores with a few extremely low scores,the mean will be the lowest,the median in the middle,and the mode the highest.

Question 4

A researcher wants to express the middle of a distribution of numbers whereby half the cases are higher and half the cases are lower than the middle value.What statistical measure should the researcher use?

A)Mean
B)Median
C)Mode
D)Standard deviation
E)Correlation

Question 5

Calculate the mean for the following six shoe sizes: 9,10,10,8,12,11.

A)8
B)8)5
C)9
D)9)5
E)10

Question 6

Describe four ways a researcher can display information about univariate statistics.

Question 7

Which neighbourhood(s)has an income distribution resembling a skewed curve?

A)Glenbrook
B)Meadowbrook
C)Elmbrook
D)Glenbrook and Elmbrook
E)Meadowbrook and Elmbrook

Question 8

Describe,as simply as possible,what is meant by the statement "It is statistically significant at the 0.05 level."

Question 9

What are two ways in which statistical relationships can be described? Provide an example for each one.

Question 10

Why is knowing the variability or dispersion of a variable as important as knowing its central tendency? How is variation measured?

Question 11

A "codebook" is

A)only used in existing statistics research.
B)a document that tells the researcher where variables are located in the data file and what numbers go with what variable attributes.
C)the set of instructions that tell interviewers or experimenters how to treat respondents or subjects.
D)an unnecessary part of data analysis since computers were invented.
E)a sheet of paper with a grid of 80 columns corresponding to data card columns,with rows representing an individual card.

Question 12

An elementary school teacher has three classes.She finds the following mean and standard deviations for student IQ scores. She knows she is most effective when the students all have similar IQ levels.Which class is she likely to be most effective with?

A)Class #1
B)Class #2
C)Class #3
D)Classes #1 and #2 equally
E)Classes #1 and #3 equally

Question 13

You discover that the years of marriage before a divorce for nine whites and nine nonwhites are as follows: Whites 12,1,8,9,10,17,3,6,6
Non-Whites 1,9,15,18,11,13,14,7,3
Which statement about this data is true?

A)There is no difference in the range for the two groups.
B)The median years of marriage prior to divorce is three years longer for whites than nonwhites.
C)The mean years of marriage prior to divorce are the same for both groups.
D)On average,nonwhites stay married longer prior to divorce than whites.
E)The mode for the two groups is the same.

Question 14

Which neighbourhood(s)has an income distribution resembling a normal curve?

A)Glenbrook
B)Meadowbrook
C)Elmbrook
D)Glenbrook and Elmbrook
E)Meadowbrook and Elmbrook

Question 15

Calculate the median for the following six shoe sizes: 9,10,10,8,12,11.

A)8
B)8)5
C)9
D)9)5
E)10

Question 16

Explain how a researcher codes,enters,and cleans data so that it can be used for statistical analysis.What procedures are involved in each of these steps?

Question 17

What are five measures of association that are useful when interpreting bivariate statistics? Describe each one and also specify which level of data each one is applicable to.

Question 18

Calculate the mode for the following six shoe sizes: 9,10,10,8,12,11.

A)8
B)8)5
C)9
D)9)5
E)10

Question 19

Describe each of the three techniques researchers use when deciding whether a relationship exists between two variables.

Question 20

Which neighbourhood(s)has the greatest differences in family income?

A)Glenbrook
B)Meadowbrook
C)Elmbrook
D)Glenbrook and Elmbrook
E)Meadowbrook and Elmbrook

Question 21

Susan weighs 140 pounds.You learn that for the women in her sorority the mean weight is 130 pounds,the median is 125,the mode is 120,and the standard deviation is 10 pounds.What is Susan's z-score in the distribution weight in the sorority?

A)zero
B)1
C)2
D)1)5
E)-1

Question 22

Which neighbourhood(s)has the highest number of families with the lowest incomes?

A)Glenbrook
B)Meadowbrook
C)Elmbrook
D)Glenbrook and Elmbrook
E)Meadowbrook and Elmbrook

Question 23

What type of relationship exists between level of violence and level of education?

A)There is no relationship.
B)There is a negative relationship.
C)There is a positive relationship.
D)There is a nonlinear relationship.
E)There is a recursive relationship.

Question 24

How many individuals with 17+ years of education engaged in a level of violence defined as medium?

A)20
B)180
C)160
D)55
E)5

Question 25

What does it mean if someone tells you that they calculated a correlation of -0.75?

A)There is a strong negative relationship between two ratio or interval level variables.
B)There is statistical independence between the two variables.
C)The variables that were used were measured at the nominal level and the relationship between them is strong.
D)The researcher looked at two ordinal-level variables and found no relationship between them.
E)The person made an error in calculation: correlations can never be negative numbers.

Question 26

If the true situation in the world is that there is no causal relationship,but a researcher states that there is a causal relationship,what error occurs?

A)Type I
B)Type II
C)Falsely reject the null hypothesis
D)Falsely accept the null hypothesis
E)A and C

Question 27

Which neighbourhood(s)has the 50th percentile of education as a college education?

A)Glenbrook
B)Meadowbrook
C)Elmbrook
D)Glenbrook and Elmbrook
E)Meadowbrook and Elmbrook

Question 28

Among the violent offenders in the following table,what is the ratio of males to females?

A)1 male to 1 female,or 1:1
B)2 males to 3 females,or 2:3
C)4 males to 1 female,or 4:1
D)2 males to 1 female,or 2:1

Question 29

Which neighbourhood(s)has a small number of very high income people and a greater proportion of families earning $26,000?

A)Glenbrook
B)Meadowbrook
C)Elmbrook
D)Glenbrook and Elmbrook
E)Meadowbrook and Elmbrook

Question 30

Which neighbourhood(s)has the smallest differences in family income?

A)Glenbrook
B)Meadowbrook
C)Elmbrook
D)Glenbrook and Elmbrook
E)Meadowbrook and Elmbrook

Question 31

Your research assistant finds a positive relationship between mother's education and income at age 40.The relationship remained after controlling for the family's social class.This means that

A)social class is really the causal variable.
B)social class does not have an impact on the relationship.
C)the original relationship is spurious.
D)income is caused by neither the mother's education level nor her social class.
E)the mother's education has no effect on her future income.

Question 32

Which of the following research topics is an example of multivariate analysis?

A)An analysis of the ages of all women who are corporate executives
B)An analysis of the relationship between age,sex,and type of restaurant frequented in Saskatoon
C)An analysis of the relationship between undergraduate majors and level of position held in a major corporation
D)An analysis of the relationship between type of offense and length of prison sentence
E)The relationship between socio-economic status and annual income

Question 33

Which of the following tells how well a set of variables explains a dependent variable?

A)Z-score
B)Standard deviation
C)Chi-square
D)R-squared
E)Gamma

Question 34

Which neighbourhood(s)has the greatest variation in years of education?

A)Glenbrook
B)Meadowbrook
C)Elmbrook
D)Glenbrook and Elmbrook
E)Meadowbrook and Elmbrook

Question 35

If the true situation in the world is that there is a causal relationship,but a researcher states that there is no causal relationship,what error occurs?

A)Type I
B)Type II
C)Falsely reject the null hypothesis
D)Falsely accept the null hypothesis
E)B and D

Question 36

Fatima Fashionista wears a size 2 blazer.She went to her favourite clothing store and found that the mean size of the store's stock of blazers is a size 10 with a standard deviation of 4 sizes.What is her z-score in the distribution of the store's blazers?

A)zero
B)1
C)2
D)-1.5
E)-2

Question 37

Connie and Tom both received 75 percent on the social research methods final exam.The mean for all women who took the test was 80 percent and the median was 75 percent with a standard deviation of 5 percent.The mean and median for the men was 65 percent with a standard deviation of 10 percent.What is the z-score for Connie and Tom relative to those of their own sex?

A)Connie's z-score is +1 and Tom's is -2,so Connie did better.
B)Connie's z-score is -1 and Tom's is +1,so Tom did better.
C)Both Connie and Tom have the same z-score,1.
D)Connie's z-score is 0 and Tom's is +1,so he did worse.
E)Connie's z-score is -2 and Tom's is -1,so Tom did better.

Question 38

In which neighbourhood(s)do half the families have incomes of $28,000 or more?

A)Glenbrook
B)Meadowbrook
C)Elmbrook
D)Glenbrook and Elmbrook
E)Meadowbrook and Elmbrook

Question 39

All the sales representatives at Acadia Insurance are female,and their mean annual salary is $60,000 with a standard deviation of $5000.All customer account managers are male and they have a mean salary of $80,000 with a standard deviation of $15,000.Heather knows she is one standard deviation above the mean of the sales representatives.She wants to transfer to become the first female customer account manager and will begin at her same salary.After she transfers,compared to the customer account managers her salary will be at what z-score?

A)-2
B)-1
C)0
D)+1
E)+2

Question 40

What feature of experimental research allows it to demonstrate causality without control variables?

A)Experimental researchers cannot demonstrate causality without control variables.
B)Experimental researchers eliminate alternative explanations by choosing a research design that physically controls potential alternative explanations for results.
C)Experimental researchers eliminate alternative explanations by testing temporal order and association.
D)Experimental researchers can be assured causal relationships are not spurious because they use random sampling procedures.
E)Experimental researchers make use of partials instead of control variables in order to demonstrate causality.

Question 41

Define the following:
codebook

Question 42

Usually,when constructing a scattergram,the __________ goes on the X axis and the __________ goes on the Y axis.

A)independent variable; dependent variable
B)dependent variable; independent variable
C)form; direction
D)direction; form
E)control variable; causal variable

Question 43

Define the following:
bivariate statistics

Question 44

What do social scientific researchers call the process of checking the categories of all variables for impossible codes?

A)Possible code cleaning
B)Contingency cleaning
C)Wild code checking
D)A and B
E)A and C

Question 45

Define the following:
code sheet

Question 46

Which measure of association can never take a negative value?

A)Lambda
B)Gamma
C)Tau
D)Rho
E)Chi-square

Question 47

Define the following:
cell of a table

Question 48

Which of the following statements about the following numbers is true: 2,7,6,9,4,6,2,8?

A)The median is a whole number
B)The distribution is bimodal
C)The mean is higher than half of the numbers in the distribution
D)A and B
E)A and C

Question 49

Define the following:
contingency cleaning

Question 50

Define the following:
body of a table

Question 51

Define the following:
control variable

Question 52

Define the following:
bar chart

Question 53

Define the following:
bimodal

Question 54

Professor Quincy Quacker found a statistically significant relationship between variable X and variable Y.It is statistically significant at the 0.05 level.What does this mean?

A)There are 95 chances in 100 that the results are true.
B)There is a 5 percent chance that the results are true.
C)If 100 samples were drawn,results like these could be obtained by pure random chance 10 percent of the time.
D)A person could be 95 percent sure that the results of the study were a reflection of the population if random sampling was used.
E)A and D.

Question 55

Define the following:
correlation

Question 56

Which of the following is an example of a multivariate statistic?

A)Multiple regression
B)Standard deviation
C)Z-score
D)Chi-square
E)Age

Question 57

What should you see in a scattergram if the bivariate relationship depicted in the scattergram is characterized by a high level of precision?

A)Data points widely spread around the line
B)Data points hugging the line
C)A low number of data points
D)A high number of data points
E)A curved line

Question 58

Define the following:
contingency table

Question 59

Which of the following cannot be learned from a scattergram?

A)Form
B)Intensity
C)Direction
D)Precision
E)None of the above

Question 60

After administering a measure for Generalized Anxiety Disorder to eight-hundred randomly sampled participants,Dr.Winkler calculated the mean GAD-scores for men and for women.Dr.Winkler found the mean GAD-score for men was higher than the mean-GAD score for women,but concluded that the difference between the two means was not statistically significant.What is the difference between the sample means tell us about the relationship between the mean GAD-scores of men and women in the population?

A)The mean GAD-score for men is higher than the mean GAD-score for women in the population
B)The mean GAD-score for women is higher than the mean GAD-score for men in the population
C)There is no difference between the mean GAD-score for men and the mean GAD-score for women in the population
D)The mean GAD-score for men is equal to or higher than the mean GAD-score for women in the population,but not lower
E)The difference in sample means tells us nothing about the difference in population means

Question 61

Define the following:
histogram

Question 62

Define the following:
percentile

Question 63

Define the following:
partials

Question 64

Define the following:
pie chart

Question 65

Define the following:
mean

Question 66

Define the following:
median

Question 67

Define the following:
cross-tabulation

Question 68

Define the following:
multimodal

Question 69

Define the following:
direct entry method

Question 70

Define the following:
mode

Question 71

Define the following:
marginals

Question 72

Define the following:
descriptive statistics

Question 73

Define the following:
independence

Question 74

Define the following:
frequency distribution

Question 75

Define the following:
normal distribution

Question 76

Define the following:
linear relationship

Question 77

Define the following:
level of statistical significance

Question 78

Define the following:
curvilinear relationship

Question 79

Define the following:
multiple regression

Question 80

Define the following:
measure of association