2. Educational Research deals with
two kinds of statistical data,
descriptive data and statistical data
(Mathews & Ross,2010).
3. Statistical Analysis is a series of
techniques in presenting the findings for
analysis and interpretation. This was
done to explore and address the research
questions that have been posed for
interpretation.
5. A hypothesis is a prediction or guess of the
relation that exists among variables being
investigated
(Wonnacott&Wonnacott,1990).
A hypothesis must be stated so that it is
capable of being either refuted or confirmed. The
result will answer relationships that exist among
variables.
6. In the previous cited example, on students who
would frequent visit the library, a perception may
be formed.
As study wanted to find out whether, the
students who are “frequent library users” or “
seldom library users” will differ their
academic performances.
7. According to Wonnacott (1990), we
usually settle this argument by
constructing a 95% confidence interval.
In general, any hypothesis that lies
outside the confidence may be judged
implausible, that is, it can be rejected.
8. On the other hand, any hypothesis that lies within
the confidence interval maybe judged plausible or
acceptable. In conforming to the tradition, we
usually speak of testing at an error of 5%.
9. The hypothesis, according to the author
(Wonnacoot &Wonnacott, 1990), is of particular
interest, it is called null hypothesis since it
represents no difference whatsoever. In rejecting
it because it lies outside the confidence level, we
establish the important claim that there was
indeed a difference between students who are
“frequent users”. The result is traditionally called
statistically significant at 5% significance level.
10. There is problem with the term “statistically
significant”. It is a technical phrase that simply means
enough data has been collected to establish that
differences do exist. It does not mean that the
difference is necessarily important. Wonnacott and
associate went on to explain that,
11. Statistically significant at 5% significance level is
the traditional phrase typically encountered in the
scientific literature. It means exactly the same
things as our statistically discernible at 5% error
level.
12. If a 5% level of significance is being used, it would
be natural to speak of the hypothesis being tested
at a 5% confidence level. Now, to return to our
example, let us formally conclude that 5% level of
significance, we can reject the hypothesis of no
difference.
13. In other words, we have collected enough
evidence so that we can see a difference in
academic performance between “frequent library
users” and “ seldom library users”. This means that
the result is statistically different.
14. In print data of commonly used statistical package,
the decision criteria for accepting or rejecting
hypothesis is on the computed p-value
(significance level). The p-value summarizes
clearly how much agreement there is between the
data or null hypothesis (Ho). The p-value is an
excellent way to summarize what data says about
the credibility of Ho.
15. The statistical test is used to determine whether or
not a hypothesis is correct by telling the researcher
how likely it is that the results of an experiment are
due to chance alone. Generally, a null hypothesis is
set up stating that there is no difference between the
control and experimental samples. The data are than
collected and analyzed, and the null hypothesis is
either accepted or rejected.