>>Return to Tell Me About Statistics!
The chi-square test, which medical practitioners often use in the literature, is also called the “test of independence.” The term “independence” here simply implies, “independence=no relationship” or “non-independence=some relationship.”
What is the chi -square test?
Table 1 shows the results of examining the efficacy of new drug X in the young, middle-aged, and old age groups.
[Table 1] Effective number of new drug X by age group
This table shows that the number of people for whom the new Drug X was judged to be effective was largest among the young age group, followed by the middle-aged and older age groups, in that order.
However, based on this table alone, it is not possible to conclude that “new drug X was the most effective in young people, followed by middle-aged and elderly people.” This is because no people were judged as invalid in this table.
Table 2 shows the relationship between age groups and validity.
[Table 2] Number/percentage of valid and invalid of new drugs
The proportion of people for whom the drug was judged to be effective was 75%, 67%, and 33% in the young, middle-aged, and the elderly groups, respectively. Younger age groups tended to have a higher probability for effectiveness. The percentages of people judged to be invalid were 25% of the young, 33% of the middle-aged, and 67% of the elderly groups. Older age groups tended to have a higher probability for invalidity.
When the percentage of people judged as valid or invalid in a certain age group was high, we interpreted that the age group and valid/invalid items were related.
What about the presence or absence of a relationship between age groups in the population and valid/invalid items? The chi-square test is used to examine this. In Vol. 22, we will explain “expected frequency” and “chi-square value” in order to understand the chi-square test.
>>Return to Tell Me About Statistics!
Comments are closed