Scribd
Upload
45 views3 pages

Week 3-Chi-Square

Crosstabulation and chi-squared tests are used to analyze relationships between categorical variables. Crosstabulation compares variables by displaying counts in a contingency table with rows and columns. Chi-squared tests whether observed counts differ significantly from expected counts, indicating a relationship between variables. A significant chi-squared result occurs when the p-value is less than 0.05, meaning there is a less than 5% probability the results occurred by chance. While a significant result suggests a relationship, more analysis is needed to describe the nature of the relationship.

Uploaded by

Joanna Burdeos
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
45 views3 pages

Week 3-Chi-Square

Crosstabulation and chi-squared tests are used to analyze relationships between categorical variables. Crosstabulation compares variables by displaying counts in a contingency table with rows and columns. Chi-squared tests whether observed counts differ significantly from expected counts, indicating a relationship between variables. A significant chi-squared result occurs when the p-value is less than 0.05, meaning there is a less than 5% probability the results occurred by chance. While a significant result suggests a relationship, more analysis is needed to describe the nature of the relationship.

Uploaded by

Joanna Burdeos
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 3

Crosstabulation and chi-squared test

SOC 195, Prof. Llangco

RECALL NOTES

💡DESCRIPTIVE  Descriptive: Describe data and distributions in standardized,


often pictorial ways (e.g., mean, standard deviation, graphs,
VS. INFERENTIAL charts, and tables)
STATISTICS  Inferential: Make decisions about how variables are related
(e.g., conclusions and generalizations)

ANALYZING RELATIONSHIPS
 Contingency tables or crosstabulations: compares
nominal or categorical variables (but can include ordinal
variables)
 N.B. tables contains counts (frequency data)
 One variable on horizontal axis
 One variable on vertical axis
 Row and column total counts known as marginals
 Use column and row percentages to look for relationships

💡CHI-SQUARE  Cross tabulations and Chi-square are tests that can be used
STATISTICS AND to look for a relationship between two variables
REPORTING  When the variables are categorical, so the data are
RESULTS nominal (or frequency)
 For example, if we wanted to look at the relationship
between gender and age
 Several different types of Chi-square, we will be using
the 2x2 Chi-square (2 columns, 2 rows)
 Inferential stats and ‘p’
 When testing if a relationship is significant
 Test how likely is the null hypothesis
 Null hypothesis is no relationship or no difference
 How likely to make type 1 error- accept as true when
false (crying wolf)
 Statistics calculate a probability of this error
 We want this to be very small
 ‘p’ value
 This is shown as e.g. “p=0.003”
 p varies from 0 to 1
 By convention we look for p<0.05
 This is called the significance level of the test
(alpha)
 1 in 20 or 5% chance or lower
o But can use p<0.01 if we want to be very sure
of our results
 I.e., Reject null hypothesis if p<0.05

 I.e. relationship is significant (we are unlikely to have


got that by chance) if p<0.05

PROCEED WITH CAUTION


 N.B. Significant does not mean necessarily true
 p = 0.05 means there is a 1 in 20 chance of being
wrong
 Need to say what the relationship is
 Statistic may be unreliable if:
 It’s 2x2 table – then use continuity correction
If any expected values are ≤ 5 (look at the table)
📌 SUMMARY

You might also like