Deck 16: Analyzing Data and Drawing Conclusions

Question 1

The analysis of data in social work research can be used to ______.

A) describe people
B) test the hypothesis in explanatory research
C) describe people and to test the hypothesis in explanatory research
D) describe people and to generalize our findings on a scientific basis

Answer 1

describe people and to test the hypothesis in explanatory research

Question 2

What step in the research process comes before the analysis of data?

A) the development of the knowledge base
B) the drawing of conclusions about the data in regard to the basic research question
C) the discussion of the implications of the data results to the improvement of social work practice
D) the discussion of the strengths and weaknesses of the study methods in regard to the study findings with recommendations on how the study methods for this study could be improved

Answer 2

the development of the knowledge base

Question 3

If you had a question on a questionnaire that asked for the respondent to mark either Male or Female for the variable of gender, you would have a variable that is measured at what level?

A) ordinal
B) nominal
C) ratio
D) interval

Answer 3

nominal

Question 4

Which of the following is a descriptive statistic?

A) the mean
B) the most often cited word that reveals one's most important need
C) the correlation coefficient
D) the t test

Question 5

The proportion is ______.

A) an example of an inferential statistic
B) the average of a variable measured at the interval level
C) the most dominant answer to a question that is measured at the ordinal level
D) the percent of people who fall into a category

Question 6

A descriptive statistic that could be calculated for a variable measured at the ordinal level but not the nominal level is the ______.

A) median
B) mode
C) standard deviation
D) frequency

Question 7

The standard deviation is a measure of ______.

A) correlation
B) dependency
C) relationship
D) variance

Question 8

Inferential statistics are used to ______.

A) describe people
B) test a hypothesis
C) select a study sample
D) determine the tool to use to measure a study variable

Question 9

What are two of the questions you must answer in order to select an inferential statistic?

A) What conclusions do you want to draw? Which variable is the dependent one?
B) At what level is each variable measured? How many variables are in the analysis?
C) What variable is the entropic variable? How many people are in the study sample?
D) Are there any variables in the analysis that are measured at the ratio level? How many people are in the study population?

Question 10

Suppose that you have the following study hypothesis: Males will report higher salaries than females when years of experience and position level are taken into consideration. What is the independent variable in this hypothesis?

A) position level
B) years of experience
C) gender
D) males

Question 11

Suppose that you have the following study hypothesis: Males will report higher salaries than females when years of experience and position level are taken into consideration. How many variables are in this hypothesis?

A) One
B) Two
C) Three
D) Four

Question 12

Which of the following is an example of the structure of your data when you are seeking a statistic for testing your study hypothesis?

A) You are examining the matched pretest and posttest scores for a group of clients.
B) You are examining the proportions of people in your study who have at least one child living in the home.
C) You are computing the mean salary of the people in your study sample.
D) You are computing the correlation coefficient between three variables.

Question 13

A correlation coefficient ______.

A) reveals the strength of the relationship between two variables.
B) will provide you with information on the proportion of clients who improved on the target behavior.
C) will provide you with information on the means and standard deviations of data for your dependent variable.
D) shows whether the study results can be generalized on a scientific basis.

Question 14

Suppose you have composed the following study hypothesis: There is a positive relationship between grades in school (measured numerically rather than by letters) and scores for self-esteem. What statistic would be appropriate for testing this hypothesis?

A) Chi-square, which examines the relationship between two variables where each is measured at the nominal level.
B) Independent t test, which examines the relationship between one dichotomous variable and one variable measured at the interval level.
C) Paired t test, which examines a set of matched scores for one group of people.
D) Pearson correlation coefficient, which examines the relationship between two variables where each is measured at the interval level.

Question 15

A perfect empirical relationship that is positive would be indicated by a correlation coefficient of ______.

A) 0.50
B) 0.05
C) 1.0
D) -1.0

Question 16

When you compute the effect size in a one-group pretest-posttest study using a score for measuring depression, an effect size of 1.0 would mean that ______.

A) the data were found to be statistically significant
B) the typical client in the study had gained on depression by an amount that represents one standard deviation of scores
C) the data were not found to be statistically significant but were found to be of practical significance
D) the data were found to be neither statistically significant nor of practical significance

Question 17

Suppose you have the following hypothesis: Scores for anxiety for a single client will be lower during the treatment period than the one score taken for the baseline. What statistic would be appropriate for testing this hypothesis?

A) The independent t test which compares the scores of two groups of people.
B) The chi-square test which examines the relationship between two variables where each is measured at the nominal level.
C) The Spearman correlation which examines the relationship between two variables where each is measured at the ordinal level.
D) The one-sample t test which compares a set of scores with a single score.

Question 18

In evaluative research, you can report that your data supported your study hypothesis when the data ______.

A) went in the hypothesized direction and were found to be statistically significant
B) went in the hypothesized direction and were found to be of practical significance
C) were of practical significance but not of statistical significance
D) supported the literature that was reviewed

Question 19

What should the study conclusions of your research report start with?

A) summary of the literature in regard to the theme of the study
B) summary of the data with regard to the major study question
C) presentation of descriptive data on the nature of the study sample
D) summary of your opinions based on your practice experience

Question 20

The dependent variable is the variable that is believed to be ______.

A) dependent upon a random sample in order that the hypothesis can be confirmed
B) the cause of the independent variable
C) caused by the independent variable
D) dependent upon the presentation of an experimental design in order to demonstrate causation

Question 21

A variable measured as a score is measured at the interval level.

Question 22

The variable of gender would normally be measured at the ordinal level.

Question 23

The value of the correlation coefficient can range from a low of 0 to a high of 1.0 and can be either positive or negative.

Question 24

The higher the value of the correlation coefficient, the stronger the relationship between the two variables in the analysis.

Question 25

The median is an inferential statistic.

Question 26

The mode is the value that occurred most frequently in an array of data.

Question 27

When you select a statistic to test your hypothesis, you need to know the level of measurement for each variable in your analysis.

Question 28

If your study variable measures opinions with answer like strongly agree, agree, disagree, and strongly disagree, you will have a variable that is measured at the nominal level rather than ordinal or interval.

Question 29

The independent t test is an example of a statistic that can be used to test a hypothesis.

Question 30

Suppose your treatment objective is to improve self-esteem for a group of clients and you measure each client one time before treatment begins and one time at the end of the treatment on a self-esteem scale that gives higher scores for higher self-esteem. Your mean pretest score was 16.8 and your mean posttest score was 21.5. Your data were subjected to the paired t test with these results: n = 21; t = 1.3; p > .05. Answer the following three questions: (1) What must you find in order to report that your data supported your study hypothesis? (2) Did your data support your hypothesis? (3) Explain why it did or did not.

Question 31

Name and briefly describe three levels of measurement. Why do you need to know the level of measurement of a variable when you are analyzing data?

Question 32

Consider the following study hypothesis: Scores on the Stress Scale have a positive relationship with hospital readmission (measured as either Yes or No) even when scores on the Social Support Scale are taken into consideration. Answer the following question: (1) How many variables are in the analysis of data for this hypothesis and what is the label you will give for each one? (2) What is the level of measurement for each of these variables? (3) What variable is the independent variable?

Question 33

Consider the following items of information that you might include in the study conclusions section of your research report: (a) the limitations of your study methodology, (b) the strengths of the methodology, (c) a summary of the results of your study in regard to the basic research question, (d) your opinions on the theme of the study, (e) the implications of the study results for social work practice. Answer each of the following questions: (1) Which of these items of information should come first in your study conclusions section of the report? (2) Which of these items is the least important for the study conclusions section of your research report? (3) What are some safeguards to consider if you decide to include this least important item?

Question 34

How are descriptive statistics and inferential statistics used differently in data analysis? What is an example of each?