Ch.15 | Test Bank – - Quantitative Data Analysis Descriptive

Chapter 15 - Quantitative data analysis: descriptive statistics

Test Bank

Type: multiple choice question

Title: Chapter 15 - Question 01

01) Which of the following is not a determinant of your choice of quantitative analytical method?

• The nature and quality of your data collection method. Your choice of research design and method directly affects your analytic choices. For example, whether you collected data via questionnaires with limited response options, via open ended questions on a survey, or an experiment, directly affects the analytic strategies available to analyze that data. In addition, if your data collection method was flawed in some way (for example, your survey questions were asked in a way that introduced a form of systematic response bias, your experimental manipulation did not work, or you have a lot of missing data), then this will affect your data analytic choices.
• The nature and quality of your data. Different types of variables, discussed below, necessitate different analytic techniques. In addition, low quality data, i.e., data that have been entered or processed incorrectly (e.g., when using an incorrect approach to combine survey items into a scale), have an effect on the outcome of statistical analysis and can hamper our understanding of the phenomenon of interest.
• The choice and execution of your sampling method. Your sampling approach can determine how representative your data is of a certain population, which can in turn affect whether you can appropriately extrapolate and generalize from your sample to the larger population of interest. In addition, your sample size can place limitations on which statistical analyses you can use and the strength of the inferences you can draw. This will be discussed further in the section on hypothesis testing in Chapter 16.

a. The choice and execution of your sampling method

b. The quality and nature of your data collection method

c. The analytical acuity of the researcher

d. The nature and quality of your data

Type: true-false

Title: Chapter 15 - Question 02

02) Missing data occur when questionnaires are not returned by the respondent.

a. True

b. False

Type: multiple choice question

Title: Chapter 15 Question 03

03) Which of the following is not a common type of variable in social science research?

• Ratio scales. A ratio scale has a true zero point, i.e., a point where there is a total absence of the construct that is being measured. Furthermore, the measurement intervals on the ratio scale are absolutely equal to each other. Length is an example of a true ratio scale. There is a true zero point, i.e., when something has no length. And the step from 1cm to 2cm is exactly the same size as the step from 14cm to 15cm. Only when a construct is measured on a ratio scale is it possible to make statements like: “Your hair is twice as long as mine.” Or “Your sister earns twice as much as I do.”
• Interval scales. These scales too have equal intervals, but they do not have a true zero point. They may in fact be measured from an arbitrary zero point. For example, the zero point of the Celsius temperature scale describes the point at which pure water freezes, while the zero point of the Fahrenheit scale describes the point at which water with a certain amount of salt concentration freezes. Values below zero can occur on an interval scale. (Note: A Kelvin temperature scale in contrast is a ratio scale as it has an absolute zero point.) Many psychological and sociological constructs are measured with interval scales as there is no clear zero point on many of them. For example, what would “zero” personality look like?
• Ordinal scales. Ordinal scales denote the relative position of a person or response on a rank-ordered selection of categories. For example, job titles may be rank ordered by their hierarchy in the organization. As such, a person’s position in the hierarchy of the organization could be measured by the job title they hold. Ordinal scales do not have a true zero point and do not have equal intervals between their categories. As such, there may be many different job titles that all denote sitting at the lower end of the organizational hierarchy. However, there is usually only one CEO. As such, these different categories on the hierarchy scale do not denote similar things and are not equally wide or big.
• Nominal scales. Variables measured on a nominal scale, or categorical variables, only denote which category a person or thing falls into without making statements of a ranked order or some form of quantitative difference between the different categories. A nominal scale often denotes a set of groups that are distinguishable on one qualitative characteristic. Usually, nominal categories are treated as mutually exclusive, meaning that a person usually only belongs to one category. For example, an item may ask respondents to indicate their religious affiliation or their marital status. One cannot be both unmarried and married at the same time. A specific case of a nominal variable is dichotomous variables, which are variables that only have two category choices. For example, “have you previously held job X? (yes/no)”. Sometimes, questions on a survey ask an individual to indicate all answers that apply to them, e.g., “tick all jobs you previously held in a provided set of job choices”. This type of question is ultimately treated as a set of dichotomous answer choices where each tick or lack of a tick in a box next to a response choice represents a ‘yes/no’ statement.

a. Nominal Scales

b. Bivariate scales

c. Ordinal Scales

d. Ratio Scales

Type: true-false

Title: Chapter 15 - Question 04

04) Ordinal variables are variables whose categories can be rank ordered, but the distances between the categories are not equal across the range.

a. True

b. False

Type: true-false

Title: Chapter 15 - Question 05

05) Dichotomous variables contain data that have only two categories (for example, sex)

a. True

b. False

Type: fill-in-blank

Title: Chapter 15 - Question 06

06) In order to gain information about the core characteristics data, researchers usually first examine a ____________ table.

Feedback: To gain more information on the core characteristics data, researchers usually first examine a frequency table.

Section reference: 15.4 Descriptive and univariate statistics

a. Frequency Table

b. Contingency Table

Type: fill-in-blank

Title: Chapter 15 - Question 07

07) Skewed distributions are distributions where two halves of the distribution are not _____________

Feedback: Skewed distributions, in contrast, are distributions in which the two halves of the distribution are not symmetrical.

Section reference: 15.4 Descriptive and univariate statistics

a. Asymmetrical

b. Symmetrical

Type: fill-in-blank

Title: Chapter 15 - Question 08

08) Where distributions are heavily skewed, there is a lack of __________ in the variables of interest.

Feedback: It is important to know what these potential deviations from the normal distribution indicate and the consequences they have for subsequent statistical analyses. When distributions are heavily skewed, there is a lack of variability in the variables of interest, i.e., everybody answers in the same way.

Section reference: 15.4 Descriptive and univariate statistics

a. Symmetry

b. Variability

Type: true-false

Title: Chapter 15 - Question 09

09) Univariate analysis refers to the analysis of more than one variable at a time.

a. True

b. False

Type: multiple choice question

Title: Chapter 15 - Question 10

05) Which of the following is not a form of average recognised in quantitative data analysis?

1. Arithmetic mean. This is the average as we understand it in everyday use—that is, we sum all the values in a distribution and then divide by the number of values. Thus, the arithmetic mean (or more simply the mean) for var00002 is 33.6, meaning that the average age of gym visitors is nearly 34 years of age. The mean should be employed only in relation to interval/ratio variables, though it is not uncommon to see it being used for ordinal variables as well.
2. Median. The median is the mid-point in a distribution of values. Whereas the mean is vulnerable to outliers (extreme values at either end of the distribution), which will exert considerable upwards or downwards pressure on the mean, by taking the mid-point of a distribution the median is not affected in this way. The median is derived by arraying all the values in a distribution from the smallest to the largest and then finding the middle point. If there is an even number of values, the median is calculated by taking the mean of the two middle numbers of the distribution. In the case of var00002, the median is 31. This is slightly lower than the mean, in part because some consider- ably older members (especially respondents 5 and 10) inflate the mean slightly. The median can be employed in relation to both interval/ratio and ordinal variables.
3. Mode. The mode is the value that occurs most frequently in a distribution. The mode for var00002 is 28. The mode can be employed in relation to all types of variable.

a. Arithmetic mean

b. Meridian

c. Median

d. Mode

Type: true-false

Title: Chapter 15 - Question 11

11) Bivariate analysis is concerned with the analysis of two variables at a time to uncover whether or not the two variables are related.

a. True

b. False

Type: multiple choice question

Title: Chapter 15 - Question 12

12) A correlation is a ‘standardized’ what:

a. Coefficient

b. Composition

c. Coexistence

d. Covariance

Type: true-false

Title: Chapter 15 - Question 13

13) Pearson’s r is a method of examining relationships between nominal/ordinal variables.

a. True

b. False

Type: multiple choice question

Title: Chapter 15 - Question 14

14) Spearman’s rho () is which type of correlation?

a. Rank-order correlation

b. Bivariate correlation

c. Multivariate correlation

d. Univariate correlation

Type: multiple choice question

Title: Chapter 15 - Question 15

15) Which is the most accurate description of a contingency table?

a. They display multiple variables simultaneously, providing information about re-occurrence set of responses on the multiple variables

b. They display two variables simultaneously, providing information about co-occurrence of sets of responses on two variables

c. They display single variable relationships, providing information about the response to a single variable

d. None of the above