• ## Summary

Search form
• 00:05

[An Introduction to the Chi-Square Test]

• 00:11

HERSCHEL KNAPP: Hi.I'm Dr. Herschel Knapp. [Herschel Knapp, PhD, MSSW,Adjunct Lecturer, Department of Study, Universityof Southern California] This video provides an overviewof the chi-square statistic.Chi-square compares categorical variables to each otherto determine if there's a consistency within the groups.We could ask the question, is gender evenly distributedamong academic degrees?[Example of the Chi-Square Formula]

• 00:34

HERSCHEL KNAPP [continued]: So for this example, we'll survey a group of graduatesto determine if there is an equal proportionof each gender-- females and males-- within each degree--bachelor's, master's, and doctorate.The chi-square formula operates on proportions or percentages.So it will work even if the groups are unequal.In this example, we're going to have 216 females and 112 males.

• 01:00

HERSCHEL KNAPP [continued]: And it's still going to work.Now.The chi-square is going to assess these figures.And it's going to make a determination.So if among the bachelor's degreesthere's the same proportion of females and males,and among the master's degrees there'sthe same proportion of females and males,and among the doctor's degrees there's

• 01:21

HERSCHEL KNAPP [continued]: the same proportion of females and males,then the chi-square will produce a p-valuethat's greater than 0.05.Suggesting that within each degree,there is no statistically significant differencein the percentage of females and males.Alternatively, if the gender percentagesare significantly different within any of these degrees'

• 01:43

HERSCHEL KNAPP [continued]: categories, then the p-value will be less thanor equal to 0.05.So to illustrate this, here's gender with two categories--female and male.And then we have degree with three categories-- bachelor's,master's, and doctorate.It's OK that each categorical variable

• 02:03

HERSCHEL KNAPP [continued]: contains a different number of categories.Two against three-- still going to work.The chi-square is then going to comparethe two categorical variables against each otherand produce a p-value.In this case, let's say the p comes out to 0.431.We would say that since the p-value is greater than 0.05,there is no statistically significant difference

• 02:26

HERSCHEL KNAPP [continued]: in the gender proportions within each degree.And as you can see, within the bachelor'sthe percentages of male to female are about the same.Within the master's the percentages are about the same.And within the doctorates, the percentages of the genderare about the same.Let's suppose we rerun this with a different group,and this time we get a p-value of 0.016.

• 02:50

HERSCHEL KNAPP [continued]: In this case, since the p-value is less than or equal to 0.05,we'd say that there is a statistically significantdifference in the gender proportions within at least oneof the degrees.And as you can see here, each of the degreeshas their male and female gender percentages.

• 03:10

HERSCHEL KNAPP [continued]: We're seeing that they're very different from each otherwithin most of the degrees.And this is telling us that we havea statistically significant differencehere from the chi-squared test.[Conclusion]For further details on conducting the chi-square test,please download the videos at sagepub.com/knapp.

• 03:34

HERSCHEL KNAPP [continued]: For further reading please refer to Introductory statisticsusing SPSS or Practical statistics for nursing usingSPSS, both by Herschel Knapp.

### Video Info

Publisher: SAGE Publications Ltd

Publication Year: 2017

Video Type:Tutorial

Keywords: practices, strategies, and tools

### Segment Info

Segment Num.: 1

Persons Discussed:

Events Discussed:

Keywords:

## Abstract

Dr. Hershel Knapp describes the chi-square test as a way to compare categorical variables to and determine consistency among groups.