You are on page 1of 6

Vince Nixau A.

Padel April 13, 2021


MATH 89 E1A4 Midterm Activity

A. Mean, Median, Mode, and Standard Deviation


First Article: DeNise Brock’s “An investigation of the relationship between
teacher perception of school environment and school suspensions”
Since Descriptive Statistics was used in the study, it is a must to examine the
mean, median, mode, and standard deviation. Moreover, the study utilized the Likert
ranking system ((1) Almost never; (2) Sometimes; (3) As often as not; (4) Frequently;
and (5) Almost always) to measure the teachers’ perception. Using of mean, median,
mode, standard deviation determines the possible trends of responses. In the study,
there are a total of 257 participants. The participants are teachers in one of six schools
that are divided into three groups representing schools with differing student
suspension rates. The schools in Group A have a combined staff of approximately 118
teachers. The schools in Group B have 83 teachers, and the remaining is Group C. In
the perceptions of student, the results shows that Students shows respect to their
classmate in Item 1 and 17 Students show poor sportsmanship reflect a median of N=
3.00 for all groups. Groups B and C share a mode of N= 3 (As often as not) for Item
17 Students show poor sportsmanship, while Group A resulted a mode of N= 2
(sometimes). Group B has the greatest mean for Item 1 as well as 17. For Items 4, the
mode for each group is N= 4, which show that most of the teachers in each group
perceive that students respect the personal property of others frequently. The median
score for Groups A and C is N = 3.00 with Group B reflecting a median score of N =
4.00. In item 7, group A as well as C share a median of N = 2.00 (sometimes). This
perception is supported by the mode of Group A is N= 2(sometimes), Group B is N =
3 (as often as not) and Group C, N = 1 (almost never). The median for Groups A and
C is N=2 in item 9. Group B reflects a mean of N = 3.00. However, the mode for Group
A as well as B is N = 2 with Group C supporting a mode of N = 1 (almost never). This
suggests that most of the participants in the groups perceive that the students behave
respectfully toward their teachers only sometimes, while Group C perceives it almost
never happens. In Items 12, 15, 20, and 23, Group A and B share a median of N =
2.00, with Group C resulting a median of N = 3.00. Groups A and C reflect a mode of
TV = 2 (sometimes) for Items 12 and 15, while Group C yields a mode of N = 4
(frequently). In Item 20, Group A, B, and C reflect a mode of N=2 (sometimes), N=3
(as often as not), and N=1 (almost never). The three groups reflect a mode of N=2
(sometimes for Item 23. The standard deviations are shown in the table.
Item Group A Group B Group C
Mean 3.65 3.78 3.18
Median 3 3 3
1 Students treat classmates with respect.
Mode 3 3 3
SD 1.03 1.037 1.011
Mean 3.27 3.91 3.09
Students respect thepersonal property of Median 3 4 3
4
others. Mode 4 4 4
SD 1.088 0.972 1.369
Mean 2.22 2.91 1.97
Median 2 3 2
7 Students show respect for school property.
Mode 2 3 1
SD 0.942 1.092 0.993
Mean 2.19 2.86 1.91
Students behave respectfully toward their Median 2 3 2
9
teachers. Mode 2 2 1
SD 0.924 1.065 0.885
Mean 2.2 3.12 1.96
Students are disrespectful toward their Median 2 3 2
12
teachers. Mode 2 4 2
SD 0.857 0.985 0.874
Mean 2.39 3.02 2.04
Median 2 3 2
15 Students pick on other students.
Mode 2 4 2
SD 0.93 1.045 0.904
Mean 2.63 3.27 2.5
Median 3 3 3
17 Students show poor Sportsmanship.
Mode 2 3 3
SD 0.918 0.937 1.09
Mean 2.37 2.37 2.04
Students are disrespectful toward their Median 2 2 2
20
schoolmates. Mode 2 2 1
SD 1.011 1.011 0.986
Mean 2.07 2.7 1.9
Median 2 3 2
23 Students refrain from putdowns.
Mode 2 2 2
SD 0.821 1.052 0.847
B. Confidence Interval
Second Article: Ala’a Mohammad Alsarhan’s “Alternative Methods of Estimating
the Degree of Uncertainty in Student Ratings of Teaching”
The confidence interval of a sample mean is an important sample statistic that
is intended to help researchers qualify the inferences which they make about the
population from which this sample was drawn. In this study, the confidence interval is
an estimated range of values within which the true mean of a class has a 95%
probability. The confidence interval for each class was from the mean and the
variability of the ratings obtained from the students in that class and the number of
students who responded. The study utilized different treatments to construct the
confidence interval. The treatments include Bootstrap method, resampling, Non-
parametric bootstrapping (Accelerated Bias-Corrected BCa), Bayesian bootstrap. In
resampling method, the analysis was carried out using this formula: 𝑎 =
𝛼 𝛼
[1000 × 2 ] , 𝑏 = [1000 (1 − 2 )]. This show that we select the 25th and 975th values to
construct 95% confidence interval. Non-parametric bootstrapping (Accelerated Bias-
Corrected BCa). This analysis was carried out with the bootstrap library under R,
version 3.2.2 (R Core Team, 2015). They used the accelerated bias-corrected
percentile limit which performed a resampling with replacement in this analysis. In
bootstrap method, two steps were used to generate the sample size. (1) Draw weights
from a uniform Dirichlet distribution with the same dimension as the number of data
points. (2) Calculate the statistic of each of these Bayesian bootstrap samples, using
the Dirichlet, and record it. The result shows that all confidence intervals had coverage
probabilities that were lower than the expected nominal 95% except for the logit
transformation. All confidence intervals for small samples had a narrower width than
the logit transformation interval (Table D1-D4). When the sample size is more than 20,
all interval widths were relatively similar. It is expected that when the data are
symmetrically distributed, the intervals would be symmetric and centered around the
mean (Table D4-D6). The logit transformation interval was symmetrical for different
sample sizes. All other methods require large sample size to obtain symmetrical
confidence interval. All other methods require large sample size to obtain symmetrical
confidence intervals (Tables D7-D9).
References:
Alsarhan, A. M. (2017). Alternative methods of estimating the degree of uncertainty in
student ratings of teaching (Order No. 10618372). Available from Education
Database. (1975469255). Retrieved from
https://search.proquest.com/dissertations-theses/alternative-methods-
estimating-degree-uncertainty/docview/1975469255/se-2?accountid=31223
Brock, D. (2010). An investigation of the relationship between teacher perception of
school environment and school suspensions (Order No. 3425392). Available
from Education Database. (757705952). Retrieved from
https://search.proquest.com/dissertations-theses/investigation-relationship-
between-teacher/docview/757705952/se-2?accountid=31223

You might also like