# Excel How-to for Exercise 6

### Directions for Analysis using Epi Info in Exercise 6

First, open the Epi Info Analysis program. From the menu on the left side of your computer screen, select Read(Import) to open the Socializing Individuals dataset (indiv.mdb) for analysis. A pop-up box will appear. Make sure that indiv.mdb is listed under Current Project. If it is not, then click the "Change Project" button at the bottom of the screen and direct the program to indiv.mdb.

Under Data Formats, make sure that "Epi 2000" is selected. Under "Show," make sure that "All" is selected, and under "All", click on "indiv". Then click the button that says "OK". You are now ready to proceed with analysis.

1. What is the total number of men and women who were approached for an interview?

In order to answer this question, you need to look at the gender of the respondents. From the left menu, under Statistics, click on Frequencies. A pop-up menu will appear on your screen. You can answer the question by making a table of the different values for respondent's gender. This is question D11 in the questionnaire and variable "D11" in the dataset. From the pull down menu under "Frequency Of", select variable D11 and click "OK". A table will appear in the grey box on your screen. The values in the table are 1 and 2. If you refer to the questionnaire you will find that 1 means male and 2 means female.

2. What is the total number of men and women who refused to be interviewed?

For this question you will need to look at both the variable for gender of respondent (D11) and the variable for whether or not a person was willing to be interviewed (D13). From the left menu, under Statistics, click on Tables. A pop-up menu will appear on your screen. Under Exposure Variable, select variable D11 and under Outcome Variable select D13. Then click "OK" and a table will appear that you can use to answer this question. (You will need to refer to the questionnaire for a key to the numeric coding used in the dataset.)

3. What is the total number of men and women who were successfully interviewed?

Use the table you created in number 2 to answer this question.

4. What is the age distribution of the men and women who were interviewed?

For this question you will need to create a table with the age variable (D12) and the gender variable (D11); follow the instructions for number 2, but specify D12 as the exposure variable and D11 as the outcome variable.

5. By gender, what percent of patrons (socializing individuals) interviewed:

To answer this question, create a table with the "Where do you live now" variable (D14) and the gender variable (D11); follow the instructions for number 2, but specify D14 as the exposure variable and D11 as the outcome variable.

• do not live in the PPA?

• visit the site daily?

For this question you will need to create a table with the variable about how often the site is visited (D16) and the gender variable (D11); follow the instructions for number 2, but specify D16 as the exposure variable and D11 as the outcome variable.

For this question you will need to create a table with the "Ever had sex" variable (D35a) and the gender variable (D11); follow the instructions for number 2, but specify D35a as the exposure variable and D11 as the outcome variable.

• have ever used a condom?

For this question you will need to create a table with the "Ever used a condom" variable (D36a) and the gender variable (D11); follow the instructions for number 2, but specify D36a as the exposure variable and D11 as the outcome variable.

• used a condom with their most recent new sexual partner?

For this question you will need to create a table with the variable that describes use of condoms with most recent new sexual partner (D36b) and the gender variable (D11); follow the instructions for number 2, but specify D36b as the exposure variable and D11 as the outcome variable.

• had more than one sexual partner in the past 12 months?

For this question you will need to create a table with the variable that describes the number of sexual partners a person had in the past 12 months (D28) and the gender variable (D11); follow the instructions for number 2, but specify D28 as the exposure variable and D11 as the outcome variable.

• Of patrons who have had more than one sexual partner in the past 12 months and also had at least one regular sexual partner, what percentage did not use a condom the last time they had sex with a regular partner?

First, select the patrons who have had more than one sexual partner in the past 12 months and also have had at least one regular sexual partner, so that your analysis of condom use at last sex with regular partner, by gender, will be confined only to the group of people that you want.

You can select these people by choosing "Select" under "Select/If" in the left hand area of your screen. In the white box that appears, type: D28>1 AND D32=1. You are specifying that you only want people who have had more than 1 sexual partner in the past year by typing "D28>1" and people who had at least one regular partner in the past year by typing "D32=1".

Now, make a table with the variable that describes whether a person used a condom the last time they had sex with a regular partner (D33) and the gender variable (D11); specify D33 as the exposure variable and D11 as the outcome variable.

When you are finished, don't forget to "deselect" by clicking on "Cancel Select" under "Select/If". Otherwise, your analysis will continue to be restricted to the selected subgroup.

• had one or more new sexual partners in the past 12 months?

Make a table with the variable that describes number of new sexual partners in the past 12 months (D29) and the gender variable (D11); specify D29 as the exposure variable and D11 as the outcome variable.

• had one or more new sexual partners in the past four weeks?

Make a table with the variable that describes number of new sexual partners in the past 4 weeks (D27) and the gender variable (D11); specify D27 as the exposure variable and D11 as the outcome variable.

• gave or exchanged money for sex in the past 4 weeks?

Make a table with the variable for having exchanged sex for money in the past 4 weeks (D53c) and the gender variable (D11); specify D53c as the exposure variable and D11 as the outcome variable.

6. What percent of male patrons (socializing individuals) interviewed:

• had sex with a man (MSM) in the past 4 weeks?

Make a table with the variable for having had sex with a man in the past 4 weeks (D41) and the gender variable (D11); specify D41 as the exposure variable and D11 as the outcome variable.

• had a symptom of an STI in the past 4 weeks?

Here you will need to select men who had pain on urination (D39a) or unusual discharge (D39b) or sores (D39c) and count them. Select using the instructions above; this time, in the white box, type: D39a=1 OR D39b=1 OR D39c=1. The number of men who meet these criteria will be listed in the grey window on your screen after "Record count". Divide the total number of men successfully interviewed (which you calculated above in number 3) by this number.

• When you are finished, don't forget to "deselect" by clicking on "Cancel Select" under "Select/If". Otherwise, your analysis will continue to be restricted to men who had an STI symptom in the past 4 weeks!

7. Make a two by two table that examines condom use during first sex with most recent new partner among all respondents who had a new partner in the past 12 months. Compare condom use among respondents who did talk with a peer health educator with condom use among those who did not talk with an educator at the site in the past 3 months.

Restrict you analysis to respondents who had at least one new partner in the past 12 months by using "Select". Then, make a table with the variable for condom use during first sex with most recent new partner (D30) and the variable for talking with a peer health educator (D45f); specify D30 as the exposure variable and D45f as the outcome variable.

When you are finished, don't forget to "deselect" by clicking on "Cancel Select" under "Select/If".

8. By gender, what proportion of youth aged 15-19:

Restrict your analysis to respondents aged 15 to 19 by using "Select" and respondent age (D12) for the following questions: