1 Non-Parametric Tests (Chi-Square Tests) (Dr See Kin Hai) 1.

Appropriate for data (nominal) not measured using Interval or Ratio scale. 2. Two main types of chi-square test : (a) Test for Goodness of Fit for single categorical variable to assess the relationship between available data and the predicted data based on a set of predictor variables (b) Test for Independence for relationship between 2 categorical variables. Assumptions 1. Observations randomly sampled from the population. 2. Each observation should be generated only once from a different subject. 3. If sample size is small, the lowest expected f = 5 and observed f can be any value. Example: From the table below, determine which one of the three types of TV programme is favoured by 119 teenage boys and girls.

1=male, 2=female 1=soap opera, 2= crime drama, 3=neither

How to enter Data case by case

Enter the values for the 2 variables for each of the 119 cases

How to run a chi-square 2 test

How to interpret the results of 2

Coursework (Use Prac 19b.sav)

Fourth year UBD students were asked about their career plans. 32 males and 24 females responded that they would like to work in the area of clinical psychology, while 28 males and 4 females preferred the area of organizational psychology. You wish to determine whether sex differences in career preference exist.

Refer to the Tables above

How to report the results

Fishers Exact Test

6 1. Use this 2 procedure when 1 or more of the 4 cells has an expected frequency of less than 5. Example (Enter the data as previous example for the Table given below)

(Use the steps as in the previous example) How to interpret the Fishers exact test results

How to report the results

One sample Chi-square / Chi-square test for Goodness of Fit The table shows the observed and expected frequency of smiling in 80 babies. The expected frequencies were obtained from a large scale study.

Enter data in the Data View after having named the variables and remove the 2 d.p. Label the 3 categories and weight the cells or cases with Freq

How to interpret the results

How to report

McNemars Test McNemar's test is a non-parametric method used on nominal data. It is applied to 2 2 contingency tables with a dichotomous trait (yes/no etc), with matched pairs of subjects, to determine whether the row and column marginal frequencies are equal ("marginal homogeneity"). It is named after Quinn McNemar, who introduced it in 1947. A dichotomy is any splitting of a whole into exactly two non-overlapping parts, meaning it is a procedure in which a whole is divided into two parts. It is a partition of a whole (or a set) into two parts (subsets) that are:

jointly exhaustive: everything must belong to one part or the other, and mutually exclusive: nothing can belong simultaneously to both parts.

The two parts thus formed are complements. In logic, the partitions are opposites if there exists a proposition such that it holds over one and not the other. Example The table shows the number of teenage children who changed or did not change their minds about going to university after listening to a career talk favoring university education.

10 How to interpret the results

How to report the results

Coursework (Use Prac19a.sav)