Deck 4: Using Descriptive Statistics to Describe Your Study Sample

Question 1

When you report only on the number of clients served and the average time of service provided to these clients in your agency, you are conducting:

A) an evaluative study
B) an explanatory study
C) a correlational study
D) a descriptive study

Answer 1

D

Question 2

Reporting range, proportion, mean, mode, and median constitute:

A) inferential statistics
B) qualitative exploratory knowledge
C) descriptive statistics
D) correlational statistic

Answer 2

C

Question 3

The purpose of the study, to a large extent, determines the client characteristics you most need to describe. But you are normally expected to report on:

A) the sources of referral for the clients who participated in your study
B) certain basic demographic variables such as gender, age, and race.
C) the responses of the clients to the treatment
D) the relationships between the clients

Answer 3

B

Question 4

Case level data implies that:

A) you have data on each of the variables of your study for each subject in your study
B) you have information only on certain variables for few subjects only
C) you have information on all variables for few subjects only
D) you have mostly missing data

Question 5

When you code the variable of religious affiliation, and when you assign codes such as Muslims = 3, Christians = 2, and Buddhists = 3, these codes are assigned:

A) to assert that Muslims are in a higher order than Christians
B) to say that Muslims have three times the value of Buddhists
C) to distinguish the people in your study in regard to religious affiliation
D) to treat this variable as though it were measured at the interval level

Question 6

Testing a hypothesis is a domain of:

A) descriptive statistics
B) both descriptive and inferential statistics
C) qualitative analysis
D) inferential statistics

Question 7

Descriptive statistics relate to:

A) the study sample
B) the service providers
C) the community at large
D) the population from which the sample was drawn

Question 8

The need to use inferential statistics is because:

A) we want to estimate the likelihood that our data are relevant to the sample
B) we want to describe the relationships between the subjects in the sample
C) we want to estimate the likelihood that our data are relevant to people not in our study sample
D) we want to summarize the variables

Question 9

In general, with regard to parametric and nonparametric statistics:

A) parametric statistics are used when the variable is measured at the nominal level
B) non-parametric statistics are used to nominal and ordinal data
C) parametric statistics are used when the variable is not measured at the interval level
D) non-parametric statistics are used when the variable is measured at the interval level

Question 10

With regard to the pie chart:

A) it is not useful when the number of categories is small
B) it is useful when there are categories that are dramatically smaller proportion of the whole than the others
C) it is easy to make sense of a pie chart with less categories
D) it is a good visual representation when it has small slices of the pie that they cannot be seen

Question 11

The cumulative frequency is:

A) the number of people in a given category plus the number who are in the categories that come before this category in the ordinal hierarchy
B) it is the number of people who are in the categories that come before this category in the ordinal hierarchy.
C) the proportion of the total sample who are in a given category for the ordinal variable
D) the number of people in a given category plus the number who are in the categories that come before this category in the ordinal hierarchy

Question 12

A variable is something that is constant.

Question 13

It is important to know the level of measurement because different statistical tests make the assumption that the variables in the analysis are measured at a given level.

Question 14

Variables like gender and ethnicity have categories but are not recorded in any way that is numerical in nature and yet central tendencies like mean can be computed.

Question 15

You can compute the average for variables such as age when measured in actual number of years.

Question 16

A researcher cannot treat a nominal variable as though it is measured at the ordinal level.

Question 17

Descriptive statistics include frequencies, proportions, range, mean, median, and standard deviation.

Question 18

The value of p helps us determine the probability that our data can be explained just by chance.

Question 19

We have to deal with statistical significance when we do not compare our sample to other people because we are not making any inferences about the ways in which our sample is or not different from a larger population of sample from which our sample was drawn.

Question 20

One of the assumptions of parametric statistics is that the data are not normally distributed.

Question 21

Non-parametric statistics do not make the assumption that the data are normally distributed.

Question 22

In addition to reporting on frequencies and proportions, you can also report the median to a variable measured at the ordinal level.

Question 23

Percentages for nominal variables can be displayed graphically in a bar chart, which illustrates the percentages for each category of a nominal variable.

Question 24

The proportion is not the percent of people in a category of a nominal variable.

Question 25

It is customary to report detailed data in tables.

Question 26

The proportions for each category of a nominal variable cannot be displayed in a pie chart.

Question 27

The median is the last value when all values are organized from lowest to highest.

Question 28

The median is much more likely to be reported for a variable measured at the interval level than the ordinal.

Question 29

The reason for reporting median income rather than mean income is the possibility of a small number of people in the study who may have very high salaries-thus leading to reporting a higher mean salary that it actually is. Median is not affected by extreme values on either end of the distribution.

Question 30

It makes sense to display the frequencies of an ordinal variable as a histogram because it visually shows the order of the categories of the variable.

Question 31

Describe what variance is and discuss the importance of reporting standard deviation when you report the mean score on any interval level variable.

Question 32

Please explain why larger the sample size, the closer the distribution gets to being normal.

Question 33

The author of your textbook states the following: "When your data fail to be normally distributed, it will not show the pattern of the bell curve." Please explain.

Question 34

Explain the importance of normal distribution for parametric statistics.

Question 35

What is standard deviation? Please provide an example to discuss its use and relevance in human services evaluation.