Introduction
The objective of the survey conducted was to establish if there are significant differences in what kind of food an individual takes with alcohol across the categories of gender. The sample being considered consists of 15 females and 15 males and as such gender bias is eliminated because we are dealing with equal proportions of male and females. Variables of interest in this analysis are; drinkfood, a categorical variable that captures the kind of food the participants ea when drinking alcohol. The other variable is gender which is a categorical variable capturing the sex of the respondent.
The null hypothesis
H0 – There are no significant differences in the food an individual takes with alcoholic drinks across the categories of gender.
Against the alternative hypothesis
Ha – There are significant differences in the food an individual takes with alcoholic drinks across the categories of gender.
Males are more likely to take alcohol without any food while females are more likely to take alcoholic drinks with either a snack or a meal.
Gender is the independent variable and drinkfood is the dependent variable.
The figures in the crosstabulation table indicate that 8 females which constitute 53.4% either take a meal or a snack with alcoholic drinks as compared to 6 males which make up 40% who take the same combination. The 13.4% difference between the females and males who take a meal or a snack with alcoholic drinks could lead to a significant difference across the categories of gender for the variable drinkfood. A greater proportion of males 7 which make up 46.7% will take no food with alcoholic drinks, females in the same category are 5 that is 33.35%. the 13.35% difference between the females and males who take no meal with alcoholic drinks could lead to a significant difference across the categories of gender for the variable drinkfood,
Hypothesis
Females are more likely to take a meal with alcoholic drinks than males.
The percentage of females who take alcoholic drinks with a meal are 26.7% (four females)compared to 6.7% (one male)of the males.
The test used to test if there s a significant difference between the food individuals take with alcoholic foods across the categories of gender is the chi-square test. The test is insignificant p-value = 0.523 which is greater than 0.05 (significance level) and as such it can be concluded that there is no significant difference in the foods that the participants took with alcoholic drinks across the categories of gender.The Cramer's V value is 0.274 which indicates that there is a weak positive correlation between the variables gender and drinkfood. The p-value is 0.523 and this indicates that the relationship is insignificant.
The test for significance of both the Lambda and Kruskal tau tests is insignificant at 0.05 significance levels. As such we fail to reject the null hypothesis that the differences across the categories of gender for the variable drinkfood is insignificant.