MY OFFICE HOURS ARE MONDAY AND WEDNESDAY 3:30-5  AT 307K STONE BUILDING.
MARIA HAS OFFICE HOURS TUESDAY- THURSDAY 3:15-5 IN THE LRC ROOM 124 STONE.
 

We often summarize a univariate distribution with a measure of central tendency and a measure of dispersion or variation. Measures of central tendency include modes, medians, and means, while measures of variation include the index of dispersion (D), the range, the inter-quartile range (IQR), and the standard deviation. You must decide whether each variable is nominal, ordinal or interval/ratio and examine its distribution to choose the appropriate measures of central tendency and dispersion for each variable.

A normal curve is defined by a collection of mathematical criteria. Are any of your variables normally distributed? Take a look at the criteria, and take a look at the distribution of each of your variables, and you decide.
 

REVIEW THE NORMAL CURVE CRITERIA HERE:

This total assignment is worth 20 points.

Correctly following all programming information for running frequency distributions on the four variables and turning in all output = 2 points.

Although your actual output does not count very heavily, I MUST receive your output in order for you to receive credit on this assignment.

Constructing a histogram on the variable "babies" = 3 points.

Designating the measurement level of each of the four variables correctly = 4 points.

Designating the most appropriate measure of central tendency for each of the four variables = 2 points.

Designating the correct value of the chosen measure of central tendency for each variable = 2 points.

Designating the most appropriate measure of dispersion for each of the four variables = 2 points.

Designating the correct value of the chosen measure of dispersion for each variable (when available) = 2 points.

Assessing whether one's variables are normally distributed with rationale = 3 points.

1. You will run FREQUENCIES on four variables:

maeduc (highest year of school the individual's mother completed)
  marital (the individual's marital status)
  attend  (how often the individual attends religious services) and
  babies (number of household members under 6 years old)

2. You will construct a histogram on the variable "babies"

3. You will identify the measurement level of each of these four variables as they were coded and  appeared in your computer run.

4. For each of these four variables as appearing in your computer run, you will select the most appropriate measure of central tendency, and give the value of the measure of central tendency that you chose in each case.

5. For each of these four variables as appearing in your computer run, you will select the most appropriate measure of dispersion, and then give the value of the measure of dispersion that you chose in each case (when available; "D" is not available but may be named as the most appropriate measure of dispersion depending on the particular variable).

6. You will decide if any of your variables are normally distributed (and briefly describe the rational behind your decision).

As you will see in  your computer exercise, the SDA system generates MANY statistics when it runs the frequencies program. This is typical of most statistical software, such as the SPSS program. The computer program will do the calculations, but it is your job to select among all the statistics that the SDA generates.

You are the one who decides, in fact, whether the answer will be a number, or the verbal designation associated with a numeric code that is the true category value (for nominal or ordinal variables.)

• FIRST, you will access the General Social Survey dataset and the Berkeley SDA system.
• SECOND, you will run four univariate frequency distributions with accompanying statistics.
• THIRD, after you have printed all your output, you will select the measure of central tendency and measure of dispersion that is the most appropriate for each variable.
• FOURTH, You will report the value of each of these statistics.
• FIFTH,  you will construct a histogram on the variable "babies"
• SIXTH, you will decide if any of your variables are normally distributed and give the rationale behind your choices.

  Accessing an online database (the General Social Survey) and the SDA system.
  Running a univariate frequency distribution.

NEW FEATURES include:

Constructing a HISTOGRAM (on the variable "babies") and becoming familiar with the components of one graphic display.
Assessing the appropriateness of the univariate statistics provided for each of your variables.
Deciding if any of your four variables match the criteria for a normally distributed variable.
 

You will construct a histogram on the variable "babies".

I very quickly constructed my histogram in Guide 3 using the EXCEL program. However, you can just draw your histogram by hand, or use a word processor.

Please do NOT use the bar chart option in the SDA program to do your histogram. This program option is brand new and may not work correctly. Further, the picture may not even load on your computer system. It is apparent that programmers are still working on this new option.

You can use either the percentages on  "BABIES" or the frequencies for your histogram (but not both).

If you use frequencies, please present the histogram in hundreds or thousands of cases, because otherwise there will not be enough room on the page and your histogram will be difficult to read and difficult to understand.

Recall that a histogram ALSO has a title, a data source, and a total casebase. Be sure to label each category of "babies" as well and remember the missing data.
 

REMEMBER! The first step in working with data is to ALWAYS display the total frequencies on all the variables that you plan to analyze for a particular research project.
 

SDA REVIEW

Use the RIGHT toe of your mouse to click on this link:
 

http://webapp.icpsr.umich.edu/GSS/

When the menu opens on the link, click on:         Open in New Window

After you have opened the new window and reached the General Social Survey page, now click on Analyze

This is the third from the left button, at the top of the screen.
 
 

Remember: WHAT YOU SEE BELOW IS A NONWORKING COPY.  YOU MUST GO TO THE NEW PAGE THAT OPENS ON YOUR MONITOR TO ACCESS THE PROGRAM.

You can always switch back to this screen for Assignment 2 by clicking on the box at the very bottom of the monitor screen that reads "Assignment 2". Or you can print out the pages of Assignment 2.

Once again, you will bring up the "radio buttons" screen to select an analytic option, in this case:
Frequencies or crosstabulation.
 

Study: GSS 1972-2002 Cumulative Datafile Select an action: Browse codebook in this window Frequencies or crosstabulation Comparison of means Correlation matrix Comparison of correlations Multiple regression Logit/Probit List values of individual cases Recode variables (into public work area) Compute a new variable List/delete derived variables Download a customized subset

Suggestion for running analysis programs: Click the "Open Extra Codebook Window" button above. This allows you to "copy-and-paste" the names of variables you wish to analyze from the codebook window to the analysis windows.

Return to SDA Home Page

In the Study: GSS 1972-2002 Cumulative Datafile screen that opens: first click on:
 

Extra Codebook Window

to open up the codebook window. Remember to click on the "Standard Codebook" link. Then, click on :
 

Start

so that you have the SDA program window active as well.

Get familiar with the SDA variables list. If you click on the blue underlined shorthand abbreviation or mnemonic for each variable (e.g., maeduc), you will bring up the basic univariate frequency distribution for that particular variable, including missing value codes and frequencies. Note, however, that this is for all 30 years of the General Social Survey. The SDA program typically, BUT DEFINITELY NOT ALWAYS, will omit the cases that have missing values. Double check the variable values in each case. This is one reason why you must always run the univariate frequencies at the beginning of any analyses that you conduct.

Browse around to look at the specifications of the following variables:
 

maeduc	highest year of education the person's mother completed
marital	the individual's current marital status
attend	how often the person attends religious services
babies	number of household members under 6 years old

Don't confuse the number of missing cases with the "cases excluded by filter". The cases excluded by filter ar for other years than 2002. The number of missing cases for 2002 is designated by "cases with invalid codes on row variable".
 
 

You will not have to recode any of your variables and you can obtain the frequencies and statistics for each variable in just one run.

Pull up the window for the SDA Program Screen. Again for your review, below is the example screen for running frequencies or crosstabulations that appeared when you clicked the "START" button on the SDA beginning window.
 

In the   screen on your monitor, click on the radio button that says:

o Frequencies or crosstabulation

Then click on :
 

Start

 

THE BOX BELOW WILL NOT WORK TO RUN YOUR PROGRAM.  YOU MUST GO TO THE NEW PAGE THAT OPENS ON YOUR MONITOR TO RUN YOUR PROGRAM.

Remember to "turn off" the charts option by selecting "no chart".
 

In the line marked Row, you will type:             maeduc marital attend babies

these are the four variables you will use in this analysis. Separate each variable by a space.

In the box that says Selection Filter(s):  type:         year (2002)

This will select ONLY respondents from the year 2002.

Next to Percentaging: leave the checkmark in the little box to the LEFT of "Column".

Now, click on the boxes to the LEFT of:

"Statistics"                    AND
"Question text"

to be sure both of these are checked.

Leave the check in the Color coding box.

Under "Chart options" be sure to select "(no chart)".

Blink once, then click on the gray box at the bottom left that says:
 

Run the Table

Blink again. There is your output for the four variables for this exercise.

Go ahead and print all the pages from the runs on these four variables NOW so that you have them to study when you complete this assignment. They also form part of the Assignment 2 that you will turn in to me.

Good golly! Do you ever have statistics! The problem is, which statistics are the right ones for each of your four variables?

(1) Is maeduc as presented in these data nominal, ordinal, interval, or ratio? How about marital? How about attend? How about babies?

(2) Which measure of central tendency is best for each of these four variables?
What is the category label or numerical result (when appropriate and available) of this measure of central tendency for each variable?

(3) Which measure of dispersion is best for each of these four variables?
Is this measure available for each variable?
When the measure is available, what is the category value or numerical result (when appropriate)?

(4) Do any of your variables have a normal distribution? Why or why not?
 

  Here's what you turn in to me by class Wednesday September 22  (you may add a short explanation to your answer to any of these questions):

1. Your printed output for the frequency distributions on the four variables:

maeduc; marital; attend; and babies.

2. Your constructed histogram for the variable "babies."
Use ONLY the valid values of "babies". (Do, of course, note the number of missing cases for 2002.)

3. The measurement level of each of these four variables: is each variable nominal, ordinal, interval, or ratio?

Make sure that you make this decision based on how the variable categories actually appear in your computer run.

4. The best measure of central tendency to use for each of these four variables considered alone. In each case, select only one of: the mode, the median, or the mean.

5. The numeric value or verbal category label that corresponds to the best measure of central tendency that you chose for each of these four variables. (COMBINING YOUR ANSWERS TO QUESTION 4 WITH YOUR ANSWERS TO QUESTION 5 IS FINE.)

6. The best measure of dispersion or variation to use for each of these four variables considered alone. In each case, select only one of: the Index of Dispersion (D), the range, inter-quartile range, or the standard deviation.

7. The numeric value or the verbal category label of the best measure of dispersion that you chose for each of these four variables. (COMBINING YOUR ANSWERS TO QUESTION 6 WITH YOUR ANSWERS TO QUESTION 7 IS FINE.)

PLEASE NOTE: The SDA system will not calculate the Index of Dispersion so you will not find a value of it on your output. It is OK to say so and you do not need to calculate an actual value for D if you choose this option. You probably would not want to use D for a nominal variable with a lot of categories either.
 

REMEMBER #8 IMMEDIATELY BELOW:

READINGS AND ASSIGNMENTS

OVERVIEW

Susan Carol Losh September 14, 2004
This page was built with Netscape Composer
and is best viewed with Netscape Navigator
600 X 800 display resolution.
Ivan in multicolor...