Crosstabulations/Chi-square test A political analyst suggested that people with high incomes are more likely to vote for the Greens than low-income earners. He selected a random sample of voters in the electoral district of Chelteath and asked them to anonymously provide their income level and political affiliation - Observational as researcher did not manipulate the IV (income) Why were people surveyed anonymously? By assuring anonymity, the researchers attempted to encourage people to answer honestly thereby obtaining more accurate information Interpret enlarged percentage - 37.5% of low-income earners voted Liberal Report 1. It was hypothesised that people with high incomes would be more likely to vote for the Greens than low-income earners. 2. In a random sample of 121 voters in the district of Chelteath, low-income earners were much more likely to vote Green than high-income earners. While 26.4% of people with low income voted Green, only 8.2% of high-income earners voted Green. 3. A Chi-square test shows that the relationship between income and political affiliation is significant, ×2(2) = 6.48, p = . 039. 4. Contrary to expectations, in Chelteath people with high incomes are less likely to vote Green than people with low incomes. <- Poltical affiliation * Income Crosstabulation Income Low Hgh Total Labor Count % within Intoms 36.1% 25 24 50 49.0% 41.3% Liberal Count 2 21 49 % within Income 42.9% 37.5% 39.7% Green Count % within Insome 19 4 23 26.4% 0.2% 19.0% Count 72 49 121 % within Income 100.0% 100.0% 100.0% Chi-Square Tests Pearson Chi-Square 6.475 Likelihood Ratio 7.064 2 029 Value 2 .039 Linear-by-Linear 5.051 1 .025 Association N of Vaid Cases 121 Asymp. Sig (2-sided) a. 0 cells (0.0%) have expected countless than 5, The minimum avnartar enuint le 0 31 7. Confounding factors - There are two confounding factors in the study: the effect of the movie and the effect of the time slot. You only need to explain one of these. For example: 1. Movie effect: Satisfaction with the movie experience may depend on the movie that the patrons saw rather than the chocolates. There were two groups of people in this study - those who were given chocolates and those who were not given chocolates. But the chocolates were not the only thing which differed systematically between the two groups of people. Everyone who was given chocolates saw the film 'Falling for You", everyone who was not given chocolates saw the film 'Godzilla returns to the planet of the Apes". The study design makes it impossible to tell whether the former group were happier because they got chocolates or because they saw a more enjoyable movie. 2. Time slot: Satisfaction with the movie experience may depend on the time that the patrons attended rather than the chocolates. The chocolates were only provided for the 6pm viewing of the movie. We cannot tell if the chocolates were happier because they got chocolates or because they attended this movie timeslot. Interpreting p-value (Correlations) Q. If the number of cigarettes smoked