CHI SQUARE RESULTS ANALYSIS

Chi Square results Analysis

Chi Square results Analysis

Introduction

The chi-square test for independence is a significance test of the relationship between categorical variables. This test is sometimes known as the “Pearson's chi-square” in honor of its developer, Karl Pearson. The purpose of this paper is to determine if the distribution of one variable is contingent on the second variable. Specifically, this paper will examine if the presence of social problems is related to whether or not a student dropped out of school (Moore 2007).

Table 1 is a contingency table showing whether drop out is contingent on the presence of social problems. As can be seen in Table 1, 5 of the 78 participants have dropped out without having any social problem (6.8%), compared with 5 of the 10 participants (50%) have dropped out with presence of social problems. Therefore, in this case, there is an association between the drop out behavior and presence of social problem. A key question is whether this association in the sample justifies the conclusion that there is an association in the population (Everitt, 1977).

Step 1

The table (observed values) is

Status

DROPOUT=0

DROPOUT=1

Total

SOCPROB=0

73

5

78

SOCPROB=1

5

5

10

Total

78

10

88

The chi-square test for independence, as applied to this issue, tests the null hypothesis that the drop out behavior behaviour is independent of the presence of social problem. Another way of stating this null hypothesis is that there is no association between the categorical variables of drop out and presence of social problem. If the null hypothesis is rejected, then one can conclude that there is an association in the population.

Assumptions

One has a single random sample, and this sample is cross-categorized by two variables each with 2 or more categories. Therefore, participants in this study were randomly selected and data was collected or two categories i.e. presence of social problem and drop out from college. Another key assumption of the test is that each observation is independent of each other observation. In general, this assumption is met if each participant in the experiment adds 1 to the frequency count of only one cell of the experiment. If this is the case, then the total frequency count will equal the total number of participants. Moreover, to fulfill these assumption participants must not discuss their possible responses before making them. This study fulfills this assumption because each participant in the experiment adds 1 to the frequency count. In addition, we also assumed that no discussions ...