Deck 5: Analyzing Data With Pretest and Posttest Measurements of One Group

Question 1

When you conduct a study using the one group pretest-posttest design and if you did not collect data in a manner to match each person's pretest scores and posttest scores, you may be able to complete your data analysis with:

A) the Independent samples t Test
B) pearson's product moment correlation
C) the dependent samples t Test
D) the one-sample t Test

Answer 1

D

Question 2

A human service professional is interested in studying whether EMDR (Eye Movement Desensitization and Reprocessing Therapy) is effective in reducing trauma-related symptoms among returning veterans from active combat duty in Afghanistan. Reduction in the scores on the PTSD scale would indicate the therapeutic modality's effectiveness in reducing trauma related symptoms. Therefore, in this study, the professional would:

A) examine the data to see whether there is a correlation between the pretest and the posttest scores on the PTSD scale
B) examine the data to see if there is statistical significance
C) examine the data to see whether there is statistical significance of the mean difference and whether the posttest scores is higher than the mean of the pretest
D) examine the data to compute the average of the posttest score only

Answer 2

C

Question 3

When you have pretest and posttest measurements on a dichotomous, nominal variable for a group of clients, you will use:

A) the one-sample t Test
B) the paired samples t Test
C) the dependent samples t Test
D) the binomial test.

Answer 3

D

Question 4

A non-directional hypothesis implies:

A) that the results will fall in one direction
B) that the results will fall in one of the two tails of the distribution
C) that the results will fall into either of the two tails of the distribution
D) that the results will fall in neither of the two tails of the distribution

Question 5

In human services, irrespective of the kind of t Test you use in analyzing data, your dependent variable must be measured at the:

A) nominal level
B) ordinal level
C) interval level
D) nominal and ordinal levels

Question 6

In the report that contains an evaluative study's findings, it is important to provide information pertaining to:

A) practical significance
B) statistical significance
C) statistical significance and Practical significance
D) details about the data

Question 7

The significance of the p value as the measure for statistical significance resides in the fact that:

A) it reveals the fractional equivalent of the number of times in 100 that your results would occur by chance.
B) a p value of .01 means your data would occur by chance once in 100
C) this result would be rendered statistically insignificant at a p value of .01
D) all the above statements in a-c are true and that there is evidence in support of the research hypothesis

Question 8

With regard to the value of P, it is listed in the Excel and SPSS data analysis programs as:

A) "p" in both the programs
B) "Sig (2-tailed)" in Excel
C) SPSS file lists the value of p as "p"
D) Excel file lists the value of p as "p", but SPSS reports this value in the column labeled "Sig (2-tailed)"

Question 9

In determining the relevance of the findings of our study for professional human service practice, two important factors must be considered. They are:

A) effect size and sample size
B) the extent to which clients achieved a gain that moved them from one threshold of functioning to another
C) sample size only
D) effect size and the extent to which clients achieved a gain that moved them from one threshold of functioning to another

Question 10

An example of a movement from one threshold of functioning to another level of functioning that contributes to the conclusion of practical significance for the effectiveness of the treatment is:

A) when your clients' mean self-esteem score moved from low self-esteem to moderate self-esteem
B) when your clients' mean self-esteem score moved from high self-esteem to low self-esteem
C) when your clients' mean score remains the same
D) when you switched treatment and the client's self-esteem score decreased

Question 11

The effect size is:

A) the effect of the client's attitude to treatment
B) the amount of gain in the agency's funding allocation
C) the amount of gain between the pretest and the posttest scores measured in standard deviations
D) mean difference

Question 12

When you have an effect size of 1.2 from an analysis using one group pretest and posttest, it means:

A) That the posttest scores are slightly more than two standard deviations better than the pretest scores.
B) That the posttest scores are slightly more than a standard deviation better than the pretest scores.
C) That the posttest scores are the same as the pretest scores.
D) That the posttest scores are slightly less than the pretest scores.

Question 13

When you use one sample t Test using Excel, to make sense of the output to determine the statistical significance of the gain, you will look at:

A) mean Score and effect size
B) effect size and the value of t
C) the value of t
D) mean Score, Effect size, the value of t, and the value of p

Question 14

When you use SPSS to compute t using the one-sample t Test, the mean pretest score we are using for comparison is the same as:

A) mean posttest score
B) different from the "Test value" required by SPSS
C) same as the threshold score used in Excel, but different from the "Test value" required by SPSS
D) same as the threshold value used in Excel and the "Test value" required by SPSS

Question 15

You compute the effect size by:

A) finding the difference between the mean posttest score and the mean pretest score
B) dividing the difference between the mean posttest score and the mean pretest score by the standard deviation of the posttest scores
C) subtracting the mean pretest score by the standard deviation of the posttest scores
D) adding the mean pretest score to the standard deviation of the posttest scores

Question 16

According to Cohen, the effect size of 1.0 is considered to be:

A) a small effect size
B) a medium effect size
C) a large effect size
D) an insignificant effect size

Question 17

You can measure client success for variables of nominal dichotomous level:

A) through any of the t Tests
B) by comparing an array of data to a threshold proportion
C) by calculating the proportion of successful students in the pretest period
D) by calculating the proportion of successful students in the posttest period

Question 18

The task of statistical analysis by conducting a binomial test is to:

A) to determine whether the improvement in the proportion is statistically significant
B) that the improvement cannot be easily explained by chance
C) that there is practical improvement
D) to determine whether the improvement in the proportion is statistically significant and that the improvement cannot be easily explained by chance

Question 19

The Excel file for the binomial test included in this text requires which of the following to conduct the test?

A) the proportion that serves as the threshold for comparison
B) the number of people who were successful during the posttest
C) the total number of people in the posttest data
D) a, b, & c

Question 20

The hypothesis that "liberal and non-liberal students will have different levels of aptitude for success" indicate that this is a directional hypothesis.

Question 21

In a directional hypothesis, the one-tailed t Test is used.

Question 22

With regard to testing directional and non-directional hypothesis, SPSS has the one-tailed test as the default.

Question 23

You cannot have statistical significance without practical significance.

Question 24

Examining practical significance helps in professional decision-making regarding the use and effectiveness of an intervention.

Question 25

It is important to note that there are various forms of computing the effect size and the effect size is computed differently for the different forms of the t Test.

Question 26

Using the data from Cohen listed in your text, for a one-group pretest-posttest design, an effect size of 0 would mean that there was no gain from the pretest scores to posttest scores.

Question 27

In SPSS, the output for the t Tests provide, among other information, the mean difference, the standard error of the mean, degrees of freedom, and the 95% confidence level. Most importantly, the value of t is given under the heading "t," and the value of p is given under the heading of "Sig (2-tailed)." Even if the value of is a negative value, in the report of the analysis, the negative sign is ignored, reporting the t-statistic as a number without any negative sign in front of it.

Question 28

In examining for practical significance in the output of your data analysis for one sample t Test, you may examine the difference between the mean posttest score and the threshold score in addition to examining the effect size.

Question 29

Figures computed in SPSS are always identical to those in Excel.

Question 30

In reporting the findings of your analysis for the t Tests, it is important to report the sample size, the mean posttest score, the value of t, and the value of p.

Question 31

Explain the significance of the p value reported as "p < .01" as a result of a binomial test.

Question 32

Explain how you will handle instances when your sample size is so small that only a large magnitude of gain will show up as statistically significant?

Question 33

Explain the importance of Binomial test for the posttest only design in human services evaluation.

Question 34

Explain what "Directional" and "Non-Directional" hypotheses mean. Provide examples.