DATA EVERYWHERE 2

Data Analysis in Educational Assessment

Educational assessment refers to the process of collecting data about what students have
learned within their learning environments. The process is not complete, neither does it hold
value if the data gathered is not analyzed and used to make instructional decisions. Instructors
use educational assessments to collect data that informs their instruction. The data is also used to
make administrative decisions. This may include buying more resources to enhance the
instructional process. At a national and state level, the data may be used to identify challenges
such as lack of resources and understaffing, and then necessary action is taken. In this paper, I
have identified the three measures of central tendency and described how each is useful in
helping an instructor make informed decisions pertaining to their instruction.

Mean, Median, and Mode
Fifteen students took the test and the results were as follows;
88,43,75,96,90,63,85,87,91,75,96,52,100,63, and 75. I will use this data to calculate the mean,
mode, and median. The mean is arrived at by adding all the scores and then dividing by the total
number of students (Lefrançois, 2013). In this case, the total scores for all 15 students are 1175.
The mean is
1179/15= 78.6
The median is a measure of central tendency that is arrived at by ranking all scores and
finding the midpoint (Lefrançois, 2013). In this case, the median is 85.
The third measure of central tendency is the mode which refers to the most frequently
recurring score (Lefrançois, 2013). In this case, the mode is 75.
Educational Assessment

DATA EVERYWHERE 3
Based on the bar graph, most students scored 75 and above marks, with only four
scorings below this mark. Based on the bar graph on the incorrect answers, it is clear that
question 4 tested on a content area that all the students understood; none of the fifteen students
failed on this question. Other well-performed content areas include those covered under question
eight, which had only one student giving the incorrect answer, and questions three and five,
which both had two students who failed to answer them correctly. Further, based on the bar
graph, it is evident that the content areas covered by questions two and seven were least
understood. For instance, ten students did not answer question two correctly. Also, eight students
failed question eight. Such information tells the instructor that content areas assessed under
questions two and seven were least understood.

Applying the Data

As an instructor, I would use this data to make decisions on content areas that need to be
revisited and more time spent in helping the students understand. According to Lefrançois
(2013), the information provided by the measures of central tendency helps the instructor know
which problem was the most difficult or the easiest. This way, they are able to make informed
decisions on areas that some students need individualized instruction. For instance, while an
instructor, I would revisit the content area uncovered under questions two and seven, I would,
however, not revisit the content area covered under question four. Based on the bar graph, all the
questions gave the correct response for question four. This implies that the content area was well
covered, and instructional objectives understood.
Further, even in questions where students performed poorly, some still got the correct
answers. Thus, teaching those areas again for the whole class to understand would not benefit
students who already understood them. Thus, I would apply flexible grouping. According to

DATA EVERYWHERE 4
(Ford 2005), flexible groupings help maximize the advantages of instruction to a student. When
instruction is based on student progress, flexible grouping helps accommodate the needs of
individual students. I would thus apply flexible groupings by placing students in groups based on
the questions they got wrong. I will then revisit those specific areas with those different groups.
This will ensure that students are in groups based on their needs.
Interpreting Mean, Median, and Mode

The mean, median, and mode are different in figures. Based on the computations, the
figures are 78.6, 85, and 75, respectively. Based on the scores of all the students, the median
gives a more accurate description of the scores. According to Lefrançois (2013), the median and
the mode are less likely to be affected by extreme scores in a group of data and are thus more
accurate in showing how scores are distributed. The figures, however, do not have a wide
variation. They are similar in that they all are used in describing score distributions and are thus
useful in educational assessments.
The mean, mode, and median are useful in different instances depending on the data
required. For instance, if I wanted a more accurate description of how data is distributed, then I
would use the median. For example, the median in this set of data is 85. Eight students that are
more than half the number of students who did the test, scored 85 and above. However, when
making instructional decisions on the areas to revisit based on the scores, I would use the mode.
In this instance, the mode is 75. Four students scored below 75, with three scoring 75. The
lowest score was 43. This shows that several students need a lot of scaffolding and tailored
instruction to catch up with the rest.
When sharing this data with the learners or parents, I would emphasize on the mode. I
will tell them that most students scored 75 and above. Based on the data, there were eleven

DATA EVERYWHERE 5
students who scored 75 and above. Lefrançois (2013) states that the mode can sometimes be the
most useful measure of central tendency. Agreeably, I would focus on the mode as it gives a
truer picture of the distribution of the scores.
Conclusion

Statistics form the foundation of educational instruction. Educational data helps
instructors make decisions pertaining to the effectiveness of their instruction as well as areas that
need modification (Burrill, 2011). Statistics helps instructors to reason from and about
educational data, thus enabling them to make informed decisions that are rooted in quantitative
information about students’ progress. Understanding statistics will impact on my practice as an
educator as I will be able to collect data, analyze it and use the results to make decisions on how
to improve or alter instructional practices.

DATA EVERYWHERE 6

References

Burrill, G. (2011). The role of statistics in improving education. In Satellite conference IASE,
Statistics Education and Outreach (pp. 1-4).
Ford, M. P. (2005). Differentiation Through Flexible Grouping: Successfully Reaching All
Readers. Learning Point Associates/North Central Regional Educational Laboratory
(NCREL).
Lefrançois, G. R. (2013). Of learning and assessment. Retrieved from
https://content.ashford.edu/