Executive Summary
In a semester full of uncertainty and confusion, our team wanted to provide valuable information to the Messiah University student body that could assist them in making their study time more productive and enjoyable. With a combined nearly seven years of experience here at Messiah, our team understands how difficult it can be to find a conducive study environment on campus where they munch on some food and crunch down their homework. With this in mind, we decided to provide evidence to students that will assist in this decision.
​
Our experiment tests conditions at different times, days, and locations based on the temperature, sound level, and foot traffic. In our data collection, we varied our times between afternoon and morning, our days between weekdays and weekends, and our locations between Lottie, the Union, and the Upper Library Café. Each trial consisted of three runs, where we recorded the sound level for one minute, took one temperature reading, and counted the number of people who passed by during that one-minute run. Within each trial, we ran one run every ten minutes for a distributed sample. Then with one full trial for each combination of conditions, we had three replications of our experiment for a total of 36 trials and a total of 108 runs. Also, to ensure consistency in our data, we ran one baseline test in the lower section of the library before and after each replication. To ensure consistency in our data collection, we made sure to always collect the data at the same spot for each location so that we minimized any variability within each location.
​
While most students will prefer a quieter and less distracting environment, others might be fine with a degree of noise if it means that they have the ability to socialize while studying. But for our purposes in this experiment, we are assuming the ideal conditions to be the quietest and least distracting environment. So, with that in mind, we recommend the Library on the Weekend in the afternoon as the quietest and least distracting environment. In addition, our experiment also shows that during this time, the temperature may increase to nearly 74.0 degrees Fahrenheit so make to not dress too warmly! If you need to decide between either Lottie or the Union, our experiment shows that Lottie will be a better location if you want to socialize as it has more foot traffic than the Union but is overall a quieter location.
Individual Conclusion
Aaron Bashore
In choosing what to write about, I decided to pursue something that I had considered when we were collecting data. This is the fact that I believe our method of data collection, while very comprehensive, for the most part, ignored outliers because of the massive amount of data collected. In our sound data collection, we collected one data point every 0.2 seconds, this means that between all of our runs, we had 10,800 sound data points for each location. With this massive amount of data, a couple of high outliers were for the most part ignored because they did not hold much weight in the context of the entire test. However, for the purpose of the study and the aid of the user, these high outliers are extremely important! To a student, a study location is not very helpful if it is very quiet most of the time with the occasional loud, as that occasional loud noise can be very distracting. So, I decided to analyze the occurrences of points over 70 dB, which for reference is about as loud as a vacuum cleaner.
After sorting all the data for each location into one column, I was able to create a scatterplot with all 10,800 data points for each location. Then I set the “y” bounds to only show the sounds louder than 70dB and the plot shown below in Figure 1 is the results.
Figure 1: Plot of all data points over 70 dB
Visually from the graph, you can see that there are more Union points over 70 dB. However, Lottie has the loudest overall points with some points close to 80 dB which is close to the sound level of a diesel truck. Certainly not the ideal environment for studying! Below in Table 1 are the tabulated numbers of points over 70 dB for each location as well as the average number of seconds between each of these points. To find this number, we divided our overall seconds of data collection (2160) by the number of data points over 70 dB. Because the Library did not have any points over 70 dB, we could only conclude that there are at least 2160 seconds between two points.
Table 1: Tabulated Data for sound over 70 dB for each location
Individual Conclusion
Noah Thrush
Figure 2: Average Temperature Graph
Figure 3: Average Sound Level Graph
Figure 4: Average Foot Traffic Graph
The end goal of this project was to find the best place on campus to study while also having the option to purchase some food. In the end, I think that the average temperature, sound level, and foot traffic is a great indicator of which location would be the most beneficial to study in. These statistics show us that the Library would be the best place to study for people who like a quiet and lowkey place to study. These averages show that the library provides a warm space that is very quiet and not close to other people. Before our group performed this experiment, this is what we had expected. The library is known as a great place to study because there is not supposed to be much talking or walking around, while the Union and Lottie are known for the opposite. The Union is full of students who are hanging out or decompressing from a stressful day or week, and Lottie is a dining hall, designed to allow student to fellowship with one another and enjoy a meal. In the end, I thought that this experiment was a success and reflected the truly best place on campus to study.
At the beginning of this experiment, We all had predictions about how we thought the experiment was going to go and what the final result would be in finding what the best place to study on campus would be. Through a lot of testing and data collection, I think that it was interesting that our predictions for the most part were pretty spot on. All of us had guessed that the library would be the quietest and have the least amount of foot traffic and we were correct in guessing that. What came as a surprise to me was how close Lottie and the Student union were in most of their measurements. Below in Figure 2, we can see that the Library on average had the highest temperature, followed by the Union and then Lottie. This was most likely due to the busyness and capacity of each place. Because the Library was probably the least busy of all three locations, there was less opening and closing of doors while in the other two more people are coming in and out, meaning that there are more opening and closing doors, which would let more cold air into the Union and Lottie. In Figure 3, we can compare the average sound levels in each location. As expected, the library is the quietest location, followed by Lottie and then the Union. This was an interesting statistic because even though Lottie is the primary dining hall and usually has the most people eating there, it is not as loud as the Union. This could be due to the fact the Union is used more for people to meet up with one another and converse, while people go to Lottie just to eat. Finally, in Figure 4 we can observe the average amount of foot traffic in each location. In Lottie, we see that there is the greatest amount of foot traffic on average, followed by the Union and then the library. This statistic was also interesting to me. I would infer that this is the case since the table and seating in Lottie is much more compact, so people are prone to walk closer to where someone may be sitting, so the foot traffic count in our data collection would be higher for Lottie.
Individual Conclusion
Ethan Barnes
Figure 5: Comparing the swing terms for temperature
I put together a simple bar graph shown in Figure 5 displaying the swing terms that overall affected the confidence interval. This did a sufficient job visually representing the variations in the range of means based on different combinations of locations, times, and days seen in the x-axis. It also allows for a better view of the range of potential dB values at 95% confidence compared to just displaying the confidence intervals on the y-axis instead. Like previously mentioned, this result surprised me since I expected the climate control across all three locations to maintain a very consistent average across each trial. With that said, the spikes and random nature of the y-axis could actually have an explanation based on speculation and prior problems our group encountered.
​
Our group found that in our initial measure X vary Y experiment, the temperature collection had a major spike in the afternoon values. Originally this was due to a defective temperature probe that was replaced (as seen in our cost analysis spreadsheet), however, my speculations for Figure 5 are concerned with direct sunlight exposure. The second probe used acted as a thermocouple with ambient temperature but only toward the end of data collection, I noticed its high sensitivity to light exposure (around ±2.0ºF!). I believe that the variation seen in the swing term is due to the random contact with sunlight in each location on the table resulting in a greater standard deviation across the temperature than the other response variables. Based on these findings, future work would have to be performed to prove this new hypothesis since our group did not have time to follow-up on these claims.
While completing the statistical analysis, I noticed something concerning the temperature data collection that surprised me. This finding was interesting to me due to the particular clash with my initial hypothesis while checking if the results made sense. Because the locations our group tested appeared to be most dependent on the overall environmental changes dealing with people and ambient noise, I expected to see that response variable display greater variations. Meaning, I hypothesized that the greater standard deviation of data would occur for sound as people are walking around, coffee machines are generating noise and random laughter occurs. To my surprise, the temperature generated the greatest variation in terms of the confidence interval at 95%.
Closing Remarks
Through all of the data collection and breakdown, this project has shown us that the library is truly the best place to study. All things including sound, foot traffic, and temperature all point to the library having the ideal study environment for any student that is looking to get some work done as well. This is assuming that there is available seating around the library café and the limited options will meet one’s snacking needs. Otherwise, the second-best choice would be Lottie to grab a bite but get some work done. This experiment has been very beneficial to the student body, because knowing where the best place to study can allow for learning to flourish and for grades to skyrocket.