Transcript Education 793 Class Notes

Education 793 Class Notes
Cross-tabulations
Presentation 3
Review Descriptive Statistics
• Descriptive Statistics we have covered
– Central Tendency
– Measures of Spread
• Suppose you have nominal data.
1=Catholic, 2=Jewish, 3=Muslim,
4=Hindu, 5=Protestant, 6=Buddhist
• Is it meaningful to report a mean as a
descriptive measure?
2
One-dimensional: Frequency
Tables
• It is more meaningful to know the percentage
of each type. This process divides the sample
into subgroups for further study
• Sample Size=100
– 40% Protestant
– 18% Catholic
– 10% Jewish
– 20% Muslim
– 6 % Hindu
– 6% Buddhist
3
Two-dimensional: Crosstabs
• With two nominal/ordinal variables we can
examine the two, one-dimensional frequency
tables and we can examine the joint
contingency table to look at the relationship
between the two variables.
• Example: From the 1986 CIRP data
– Variable 1: Students views on the
statement “Married women are best in the
home”
4
– Variable 2: Students’ sex
One-dimensional Tables
5
Information to Glean
• What do we learn from the onedimensional tables?
• What types of questions should we
raise?
6
Two-dimensional Table
Column Percents
VIEW8612
MARRIED WOMEN BEST IN HOME
SEX86
get file cirp8690.sav.
crosstab view8612 by sex86
/cells=count row column.
by
SEX86
STUDENT'S SEX
Page 1 of 1
Count |
Row Pct |MALE
FEMALE
Col Pct |
Row
|
1 |
2 | Total
VIEW8612
--------+--------+--------+
1 |
762 | 1590 | 2352
DISAGREE STRONG | 32.4 | 67.6 | 55.1
| 36.4 | 73.1 |
+--------+--------+
2 |
889 |
343 | 1232
DISAGREE SOME
| 72.2 | 27.8 | 28.9
| 42.5 | 15.8 |
+--------+--------+
3 |
343 |
170 |
513
AGREE SOME
| 66.9 | 33.1 | 12.0
| 16.4 |
7.8 |
+--------+--------+
4 |
97 |
72 |
169
AGREE STRONG
| 57.4 | 42.6 |
4.0
|
4.6 |
3.3 |
+--------+--------+
Column
2091
2175
4266
Total
49.0
51.0
100.0
7
Two-dimensional Table
Row Percents
VIEW8612
MARRIED WOMEN BEST IN HOME
by
SEX86
STUDENT'S SEX
SEX86
Page 1 of 1
Count |
Row Pct |MALE
FEMALE
Col Pct |
Row
|
1 |
2 | Total
VIEW8612
--------+--------+--------+
1 |
762 | 1590 | 2352
DISAGREE STRONG | 32.4 | 67.6 | 55.1
| 36.4 | 73.1 |
+--------+--------+
2 |
889 |
343 | 1232
DISAGREE SOME
| 72.2 | 27.8 | 28.9
| 42.5 | 15.8 |
+--------+--------+
3 |
343 |
170 |
513
AGREE SOME
| 66.9 | 33.1 | 12.0
| 16.4 |
7.8 |
+--------+--------+
4 |
97 |
72 |
169
AGREE STRONG
| 57.4 | 42.6 |
4.0
|
4.6 |
3.3 |
+--------+--------+
Column
2091
2175
4266
Total
49.0
51.0
100.0
8
Information to Glean
• What do we learn from the two-dimensional
table?
• Which percents do we want to read? Column
percents? Row percents?
• Defining the population of interest.
– Interested in differences between men and women
OR
– Interested in differences between ageer-ers and
disagree-ers
9
Findings
• The two-way contingency tables are the first step in
the process of discovering the factors that determine
these proportions.
• Example: Women are more likely to disagree strongly
that “Married women are best in the home” and Men
are more likely to disagree somewhat. Thus, we may
begin to infer that based on our sample, men in
general hold more conservative views on the roles of
women. (Sex matters when we consider peoples’
views on women’s roles).
10
Higher Order Tables
• You can continue to add layers to the tables
with 3 or more variables simultaneously
examined.
• Example: Sex and Views on the Roles of
Women by Political Beliefs. We will get three
separate tables, sex by views for liberals,
middle of the road, and conservatives.
11
What do we find?
12
Laptop Exercise
• Choose two nominal/ordinal variables
• Formulate a research question about
the relationship between the two
variables
• Display frequencies and crosstabs on
the two variables
• Summarize your findings
13
Next Week
• Normal Distribution: Chapter 5, p 115-136
14