Josh Pethel DA Lab 3

docx

School

Denison University *

*We aren’t endorsed by this school

Course

101

Subject

Statistics

Date

Jan 9, 2024

Type

docx

Pages

Uploaded by MagistrateAntelopeMaster4534

Lab 03 - D-Fuse Data Collection Report Josh Pethel Introduction I piloted a study to collect data about D-Fuse, more specifically the patterns behind who goes to D-Fuse (categorized by gender, athletic affiliation), how long they wait in line, if specific food takes longer, and if people decide to go to D-Fuse on their way in/way out of the Mitchell Center. While observing D-Fuse on a Tuesday afternoon, I found no obvious patterns other than ordering a smoothie increases wait times. Questions I proposed several specific questions to assess the patterns behind D-Fuse’s operations. My Questions: 1) Does ordering a smoothie have a significant impact on wait time? 2) Is there a correlation to certain times throughout the time when athletes or non-athletes go to D-Fuse? Data To answer these questions, I collected data containing columns for the time a person entered the line, the time a person left the line, the total time a person spent in line, the gender, if a person had an obvious athletic affiliation (given by if they had a team backpack, were wearing team apparel, or clearly had equipment for a certain sport like lacrosse or football) . In my dataset each row represented a specific person’s visit to D-Fuse. I chose to collect data in this way because I was able to input it quickly into excel while observing the subject’s order, and their behavior before and after entering the D-Fuse line. DFuse.csv Column Name Variable Definition Units Data Type Variable Codes and definitions Missing value codes Time Entered Line The variable represents what time the person entered the D- Fuse line Hours and minutes Quantitative Integers N/A No missing data Time Exited Line The variable represents what time the person received their Hours and minutes Quantitative Integers N/A No missing data

food and exited the line Time Spent in Line The variable represents how long it took for the person to exit the line after entering the line Minutes (no wait was hours long) Quantitative Integers N/A No missing data Athletic Affiliation The variable represents if the person is assumed to have an athletic affiliation (judged by a team backpack, team apparel, or holding equipment clearly associated with a certain sport. N/A Qualitative and nominal Yes (Y) and N (No) No missing data Gender The variable represents what assumed sex the person embodies. N/A Qualitative and nominal Male (M) and Female (F) No missing data Just entered building? The variable represents if the person entered the Mitchell Center and immediately headed to D-Fuse N/A Qualitative and nominal Yes (Y) and N (No) No missing data Exited building afterward? The variable represents if the person immediately exited the Mitchell Center after receiving their food from D-Fuse N/A Qualitative and nominal Yes (Y) and N (No) No missing data Smoothie? The variable represents if the person purchased and received a smoothie from D- Fuse. N/A Qualitative and nominal Yes (Y) and N (No) No missing data Other food? The variable represents if the N/A Qualitative and nominal Yes (Y) and N (No) No missing data

person purchased and received food/drink other than a smoothie from D-Fuse. Analysis I conducted my analysis of the D-Fuse data using RStudio. 1) The summary for the two different categories separated by the question; was a smoothie ordered? Is pretty telling when looking at the summary with respect to wait times. The entire IQR and the mean for the “no smoothie” category is between 1 and 2 minutes, whereas the minimum value for the “smoothie” category is 2 minutes, and the IQR is between 4 and 6 minutes, showing a clear difference in wait times dictated by ordering a smoothie. Therefore, the summary statistics clearly show the difference between wait times due to ordering or not ordering a smoothie. 2) When examining the median and means for entry time for athletic affiliation (or lack thereof), the values are similar, as the medians are 4:32.5 (athletic affiliation) and 4:35.5 (none), and the means are even tighter at 4:31.9 (athletic affiliation) and 4:32.5 (none). However, the means and medians hint that there may not be a specific correlation, as both of them are close to 4:30, which is the median value for the study’s range of time (between 4 and 5 pm). The 1st and 3rd quartiles are within two minutes of each other, further proving that the summary statistics suggest no correlation between athletic affiliation and entry time.

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

The boxplots above show the relationship between a smoothie being purchased (measured by yes/no) and the wait times for each order. Looking at the boxplots side by side, the entire IQR and the first of two outliers for the “No” boxplot are smaller than the minimum time value for the “Yes” boxplot, clearly showing that the “No” boxplot is much lower than the “Yes” boxplot across the board, further proven by the fact that highest outlier for the “No” plot is equal to the median of the “Yes” plot. In short, the boxplots display that there wait times for people who do not order a smoothie are much less compared to people who order a smoothie.

The dotplots above compare the time that people entered the line and are grouped by if the people had a visible athletic affiliation or not. (the y-axis is grouped into two-minute intervals) Looking at the dotplots side by side, the ranges seem to be similar, and there are no more than three people from a certain group entering the line in the same two-minute intervals. Also, both distributions for the time data seem pretty normal, as there are <20 data points for each category, so the gaps in the intervals are normal. When considering the observations, there is no visible relationship between time entering line and athletic affiliation. However, judging athletic affiliation was difficult because all I was basing off of was a piece of equipment or team apparel, so it is likely that one or more of the non-athletic affiliated subjects is actually an athlete. In summation, the statistics and graphs helped me easily generate a conclusion for question 1, and gives me a strong idea that question 2 is without a correlation. Conclusions (5 points) In summary, our analysis DID OR DID NOT answer our initial questions. Write a few sentences

stating the “take home message” from your summary statistics and visualizations. Discuss how these analyses helped (or didn’t help) to address your initial questions and fit into the broader context of campus life at Denison. In summary, our analysis did answer our initial questions. Starting with the second question, there is no rhyme or reason behind athletes or non-athletes populating D-Fuse at certain times, suggesting that there is no “peak” or “dead” time to flock to/avoid D-Fuse during. However, the data did show that there is a strong correlation between ordering a smoothie and increased wait times, suggesting that if a D-Fuse customer is in a rush, it would be conducive to skip the smoothie. In the future, we recommend repeating the same experiment during different days of the week and at different times. From a visual standpoint, football practice was occurring throughout the sample size, which may potentially result in a large quantity of players flocking D-Fuse post- practice, which would alter the results of the second question. Similarly, other teams may practice on different days, and there may also be dead periods on different weekdays due to classes running into the afternoon. The data categories seemed to fit, but could be enhanced by specifying which type of smoothie, in an effort to see if one drives the data up, or by doing an exit poll on if there is an athletic affiliation to remove observer bias (although response bias may enter the experiment).

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

Josh Pethel DA Lab 3

Related Documents