I. Introduction
The purpose of our survey was to find out about the ice cream eating habits and preferences of our families, friends, and coworkers. The population was American adults. We asked them our questions through Facebook, email, and facetoface so our results were convenience samples and voluntary response samples since some of the people asked chose not to respond.
The questions we asked were:
Question 1 (N1): How many times over a two week period do you eat ice cream?
Question 2 (N2): How many scoops of ice cream do you eat each time?
Question 3 (C2): What flavor do you eat most often? Vanilla, Chocolate, Strawberry
Question 4 (C1): Do you prefer your ice cream in a cone? yes/no
II. Looking at a Categorical Variable
Our responses to the question "What flavor do you eat most often?" are represented by the pie chart below. Our results were not surprising, chocolate was the most popular with almost half the vote (46.53%), vanilla was the second most popular (37.62%), strawberry had the fewest amount of votes (14.85%), and one person chose not to answer this portion of the survey.

To see if our results for flavor differed according to the question "do you prefer your ice cream in a cone" see the bar graph below. Our results stayed relatively the same, only the people who ate chocolate the most preferred their ice cream in a cone more often than not. One person surveyed chose not to respond to this question.

III. Looking at a Numerical Variable
The responses to the question "How many times do you eat ice cream over a 2 week period?" are shown in the histogram, boxplot, and summary statistics found below.
Our data is skewed to the right with more than half (56%) of our respondents polling that they eat ice cream once or twice in a 2 week period. The measures of center are represented by the mean (2.4), median (2), mode (1), and midrange (5). The best representation for our data is the median; the mode is too high which is common in positive skewed data.
The range in our data is 10 with a standard deviation of 2.15. Using the Range Rule of Thumb we get an estimation that is slightly higher than the actual standard deviation. The Range Rule of Thumb is 10/4= 2.5 and the actual SD is 2.15. This is not a good approximation of the standard deviation because it is too high, and the SD is already high because it is affected by the two outliers of 10 times/2 weeks.
Our boxplot shows our outliers which are 10times/2 weeks. I do not believe these to be errors it is believable for someone to eat ice cream most days of the week or even more than once in a day a few times a week, there were also 5 people who had ice cream 78 times in 2 weeks and this number is not far off from that.


Summary statistics:

IV. Looking for a Relationship Between Two Numerical Variables
The scatter plot below represents the relationship between the number of times people ate ice cream over 2 weeks versus how many scoops of ice cream they eat at a time, looking at the scatter plot it is hard to determine if there is correlation between them. By calculating the Correlation Coefficient, which is .226, we can determine that there is a statistical a slight positive relationship since it's greater than the absolute value of r of .196.

Correlation between Scoop and Times is:
0.22598359 
Already a member? Sign in.