Confidence Intervals, Effect Size, and Statistical Power

Part I: Concepts

Questions 1-8; Answer the questions in the spaces provided.

Part I: Questions 1a-1e

Fill in the highlighted blanks with the best word or words.

1-a) __ __ indicates the size of a difference between two populations and is unaffected by sample size.

1-b) Cohen's d measures the difference between means in terms of ___.

1-c) According to Cohen's conventions, a d value of __ __indicates a small effect.

1-d) A study that calculates the mean effect size from the individual effect sizes of many studies is called a(n) __ __.

1-e) Statistical power refers to the probability of successfully rejecting the _ _ hypothesis.

Part I: Questions2-5

Complete the following problems. If applicable, remember to show work in your homework document to receive partial credit.

2) List five factors that affect statistical power. For each, indicate how a researcher can leverage/manipulate that factor to increase power in a study.

3) In a public opinion poll, approximately 89% of Americans felt that having a police officer on patrol in the neighborhood made them safer than having no police officer on patrol. The margin of error reported was 4%. Construct an interval estimate using these figures. (See Example 8.1 in Nolan and Heinzen 2016 for review.)

4) For each of the following confidence levels, look up the critical z values in the z table for a two-tailed test:

5) Remembering that a meta-analysis calculates one mean effect size using the effect sizes of several studies, assume you are conducting a meta-analysis over a set of five studies. The effect sizes for each study are: d= .78, d = .49, d = .17, d = .63, d = .06

5a) Calculate the meanof the effect sizes of these studies. Answer

5b) According to Cohen's conventions, is the mean effect size you calculated small, medium, or large?

Part I: Questions6-8______

Compute effect size in the following questions:

A company decides to add a new program that prepares randomly selected sales personnel to increase their number of sales per month. The mean number of sales per month for the overall population of sales personnel at this national company is 32with a standard deviation of 4.1.

The mean number of sales per month for those who participated in the new program is 36.
Compute the effect size of the new sales program.

7) On a certain anxiety questionnaire, the general population is known to have a mean of 25 and a standard deviation of 4. A higher score represents higher levels of anxiety. Participants in a new relaxation program complete the questionnaire after completing the program and have a mean score of 24.

What is the effect size of the relaxation program?

8) A residential treatment facility tests a new group therapy for patients with self-destructive behaviors. The therapists hope to decrease scores on a measure of self-destructive behaviors that has a mean in the overall residential treatment population of 12 and a standard deviation of 1.5. The mean score for the patients after the new group therapy is 11.1.

What is the effect size of the new group therapy?

Part II&Part III

There is no new SPSS material this module/week. No questions for Part II or III

Part IV: Cumulative

Data provided below for respective questions.

Part IV:

Questions 1-4

A test designed to measure internalizing behavior traits was administered to 20 university students. The data from this test are shown in the column to the left.

(higher scores more internalizing).

• Enter theInternalizing Behaviors data into a new SPSS file, in one column.

1-a) Using SPSS, calculate descriptive statistics for this variable. Be sure to include the mean and standard deviation.

2) Create the appropriate graph in SPSS for the variable "Internalizing Behaviors."
Justify your choice of graph.

3) Using the mean and standard deviation of the internalizing behaviors data above (from your SPSS output), calculate the z score for a raw score of 45. (See last week's material for review.)

4) Using the mean and standard deviation of the internalizing behaviors data above, calculate the percentile of a raw score of 17. (See last week's material for review.)

Part IV:

Question 5a-5e_________

For the following scores, state the: a) mean; b) median; c) mode; d) range; e) standard deviation.

5-a) Mean

5-b) Median

5-c) Mode

5-d) Range

5-e) Standard Deviation (Note: Compute by hand using the formula from Module 3. Programs like SPSS or Excel use a slightly different formula that we won't use until Module 8 and will give you the wrong answer for this section.)