Is there a Significant Difference Between North Olmsted ENN Youtube Viewership in First Semester versus Second Semester?

Excitement for the academic year plays a major role in the spirit and environment of a high school campus. It can affect attendance at school-sponsored events and promote achievement. The North Olmsted Eagles News Network Youtube channel exists to inform the students of what North Olmsted High School has to offer, and the audience it supports may reflect the interest students have in school activities. When pondering school excitement, an interesting question arises on whether or not the student body shows more engagement at the beginning of the year or at the end. Comparing the views generated on ENN videos of first semester (defined for the sake of this study as the period from August to January 15th of every school) versus second semester (January 16th to June) allows us to gauge student interest.

The Study

The study compares two population: the population of youtube videos on the North Olmsted ENN channel that were posted between August and January 15th of each year (1st semester) and the population of youtube videos on the channel posted between January 16th and June. Samples of size 30 (so that Central Limit Theorem would apply) were taken of each population using a random number generator with numbers 1-572 for the total amount of uploads at the time of sampling. A playlist of all uploads provides an index (in chronological order) for each video that would be associated with the randomly generated value. Videos would be randomly selected until both samples have size of 30 without replacement. The videos provide date of upload, which was recorded in order to distinguish the two populations, and the views were recorded for each video. Once a size of 30 was reached for the first sample, the rest of the randomly generated numbers associated with that population were disregarded, while the generated numbers associated with the remaining sample was collected to a size of 30.

A two-sample t test was used to test whether there was a difference in true mean value of views from first semester ENN videos and second semester ENN videos. The two samples were from the population of ENN Youtube videos in each semester. Since the population standard deviations were not known, a t test statistic had to be used rather that a z. A two sample test allows use to compare the means of the two samples to find whether there is a difference in views by semester. A significance level of .05 was used. Since both samples were at least of sample size 30, the Central Limit Theorem applies. The samples were simple random and independent.

μ₁1

the mean number of views on North Olmsted ENN Youtube videos uploaded in the Semester 1 time period

µ₂ 2

the mean number of views on North Olmsted ENN Youtube videos uploaded in the Semester 2 time period


Test

Null hypothesis

H₀: μ₁1 - µ2₂ = 0

Alternative hypothesis

H₁: μ₁1 - µ₂2 ≠ 0

T-Value

DF

P-Value

1.74

55

0.087

 

 

 

 

BackGround Research

When I chose to conduct a study comparing Youtube viewership on North Olmsted ENN in 1st semester to 2nd semester, I realized a few interesting factors come into play. Viewership on school youtube videos potentially reflect school spirit at different times of the school year. I first researched school spirit and its effect on the student body. With a google search of “school spirit studies,” I found that New Zealand’s Ministry of Education uploaded a high school student’s study (http://www.educationalleaders.govt.nz/content/download/724/5982/file/cowan-sabbatical-05.pdf

) that explored a relationship between school spirit and success. I found most interesting the questions she asked high school students on whose responsibility school spirit is. Her respondents did believe spirit motivated students to perform well and to do so with pride. They believed the students generate this spirit through their own operative. Students must create the environment to harbor school pride. North Olmsted ENN reflects this as students display their personality to the entire student body and therefore influence their anticipation for school-sponsored events and activities.

An article by Psychology Today (https://www.psychologytoday.com/us/blog/the-campus/201101/theyve-got-the-spirit-yes-they-do) outlines a similar idea at the collegiate level. Alan Reifman argues that schools with larger athletic programs also have the greatest academic stature, which he attributes to fanbase formed by the student body in support of the school’s sports teams.


I then used google to search for successful high school news networks throughout the country. I wanted to know what aspects are implemented to North Olmsted ENN, and if they lead to high viewership.  I found https://www.broadcastingcable.com/post-type-the-wire/california-high-school-delivers-award-winning-broadcast-media-program-wirecast-171455, which describes a high school’s media broadcasting program. The school has loads of equipment, but what struck me was their variety of broadcasting forms. They cover on sight at football games and have a portable trailer, The projectors are high tech and allow for customizability. North Olmsted does not have the resources for such a program, but ENN substitutes by posting more special videos, such as sport recognition videos for every season, and advertising for school events. Accordingly, when the school has more going on, I tend to notice more views on videos.


The beginning of the school year comes with a plethora of community events such as the Homecoming dance and the highly attended football games. Returning from summer break, I would guess that more students are willing to partake in such events than they are in the spring when they are cramming for exams and are generally worn out from the school year. As I researched to find when North Olmsted High School seems to be the busiest, I realized that the school twitter (https://twitter.com/NOEagles?ref_src=twsrc%5Egoogle%7Ctwcamp%5Eserp%7Ctwgr%5Eauthor) along with the principal’s twitter (https://twitter.com/mr_weagley) have posted at a steady pace throughout the school year, suggesting that noteworthy activity at the school that may appear on ENN is close to even for both the first semester and second semester. Even then, the study may suggest a difference between ENN’s successful coverage of such events, as well as student interest in school news among the two semesters.

 

Raw Data

 

Semester 1 Semester 2
27 178
18 48
91 24
48 49
16 17
126 12
21 78
19 12
106 42
19 7
105 12
26 45
18 6
83 43
199 10
105 130
91 28
97 19
17 25
17 14
14 11
85 9
19 63
15 36
27 42
53 88
11 26
71 43
93 30
102 19

Graphs

Descriptive Statistics
Sample N Mean StDev SE Mean
Semester 1 30 58.0 46.6 8.5
Semester 2 30 38.9 38.0 6.9

The histogram for semester 1 shows that the data has a
greater range, greater mean, and greater standard
deviation than the semester 2 sample. The descriptive
statistics support this, as the average number of views
for semester 1 videos is 9.1 views greater than for
semester 2, and the standard deviation for semester 1 is

8.6 greater than semester 2. Both distributions
are right skewed, as only a few videos tend to have a
large number of views. The box plot supports this as
well, as semester 1 has a greater interquartile range
(IQR) than semester 2. Despite the graphs seeming to
imply that for some reason, semester 1 has more
variability in viewership than semester 2, the semester
2 sample actually has two outliers on the right end,
including an extreme outlier (as the video is more than
3 IQRS from the third quartile).

 

             Discussion

The study considers only the semester each video is in, and assumes that school spirit within each semester is the main factor in the amount of views on each video. However, in some school years, a special ENN video such as a music video or advertisement for an irregular event may produce outliers with large numbers of viewership. For example, the sample of second semester videos included two outliers, and both of those videos were not regular daily announcement videos. Instead, one was a “Faith Flower” music video where Ryan Cummings advertises the Valentine’s Day flower sale, and the other video, “The Art Awakens,” persuades students to take art as an elective. Other potential confounding variables include the prevalence of teachers who show ENN to their class. The teacher may contribute only one view to a video despite the fact that the entire class watched the video. If those students watched the video in  class, they may not feel inclined to rewatch it on their own time, and thus the video misses out on views. Such factors lead to data in views to be influenced by factors that don’t describe a difference between semesters.


Even so, I would feel comfortable to extrapolate the results to the intended population, that being the youtube videos uploaded by North Olmsted ENN for each semester. Since all of the videos were uploaded within the period defined of both semesters, that encapsulates all North Olmsted ENN videos. As the videos only concern residents those associated with North Olmsted, the viewers are likely to all be former or current North Olmsted residents, or relatives of North Olmsted residents. However, the amount of viewers who randomly came along North Olmsted ENN videos from recommended or by mistake, or even the amount of viewers who are from nearby cities, cannot be measured. For example, the nearby public high school Avon also has a youtube channel that posts daily announcements, and it is also called ENN. As some people from Avon may have mistook the North Olmsted ENN as their own, the results cannot be extrapolated to represent the popularity of ENN among North Olmsted residents.


To gather further insight into the change in school spirit and activity at North Olmsted High School, I hope that future studies focus on other forms of media. For example, the administrators, staff, and students at NOHS likely have a large presence on twitter and facebook. Studying activity on the North Olmsted twitter, or principal Zach Weagley’s account, may find that posts on those accounts are more plentiful at certain times of the year than in others, and that may be related to the amount of school spirit and activity at those times.


 

COnclusion

Using a two sample t hypothesis test, the mean views of samples of North Olmsted ENN youtube videos from Semester 1 and Semester 2 can be compared to find if there was a difference in mean views. I speculated that the quantity of school events, the workload students and faculty feel at different times, as well as many other factors could lead to variation in student interest in a way that would reflect in the amount of views on ENN videos. With this hypothesis test, any causation for such a difference cannot be proved, but evidence can be gathered with confidence on whether the mean number views by semester are differing. Using a .05 significance level (giving us a result we can be 95% confident with), we fail to reject the null hypothesis (that the true mean values of views on videos within those semester time periods are the same) because the p-value was greater than .05. Therefore, there is not sufficient evidence to say that the true mean number of views on 1st semester North Olmsted ENN videos is different from the true mean number of views on 2nd semester North Olmsted ENN videos.

Abstract: To gauge the level of student interest in school activities at different times of the year, the study focuses on the distribution of views on North Olmsted ENN Youtube videos, a channel that posts daily announcement to inform students of school activity. A two sample t test would be used to find if there is a difference between the mean number of views on videos of each of two samples. The two samples were obtained to investigate change in viewership between 1st semester and 2nd semester. From a pool of all videos posted on the channel from August to January 15th of any year (to represent 1st Semester), 30 videos were randomly selected without replacement using a random number generator with each number associated with a chronological index number on each video provided by the uploads playlist. The same sampling method was repeated for videos uploaded from January 16th to June (to represent 2nd Semester). Upon graphing the data obtained from the samples on histograms and boxplots, as well as obtaining descriptive statistics of mean and standard deviation, the first semester sample had a larger mean and larger variability (in terms of standard deviation and interquartile range, and range) than the second semester, and both were skewed right. The second semester had two outliers on the high end, and both of the videos that had an unusually high number of views were separate from the regular daily announcement programs that the channel normally uploads. Once the hypothesis test was run with a significance level of .05, no sufficient evidence was found to conclude that there was a difference in mean number of views for 1st semester North Olmsted ENN youtube videos and views for 2nd semester North Olmsted ENN youtube videos. Using the obtained t test statistic value of 1.74, the p-value of 0.087 was found, which meant that there was a high enough chance, at the .05 level of significance, that the samples could have produced sample means with such a difference while still coming from populations with equal true means that the alternative hypothesis (that the means were different) could not be adopted as a conclusion.