# [LON-CAPA-cvs] cvs: loncom /html/adm/help/tex Statistics_Sequence.tex

**matthew
**
lon-capa-cvs@mail.lon-capa.org

*Mon, 12 Jul 2004 19:28:22 -0000*

matthew Mon Jul 12 15:28:22 2004 EDT
Added files:
/loncom/html/adm/help/tex Statistics_Sequence.tex
Log:
Help on sequence level statistics.
Index: loncom/html/adm/help/tex/Statistics_Sequence.tex
+++ loncom/html/adm/help/tex/Statistics_Sequence.tex
\label{Statistics_Sequence}
The \textbf{Overall Problem Statistics} page displays the following
data about the \textbf{Sequences} selected for evaluation:
\begin{itemize}
\item \textbf{#Items}
The number of assessment items (problem parts) in the sequence.
\item \textbf{Score} and \textbf{Count} Statistics
For both the score, the product of the problem weight and the percent awarded,
and the count, the number of problem parts correct, the following data are
computed:
\begin{itemize}
\item \textbf{MEAN}
\item \textbf{Standard Deviation}
\item \textbf{Maximum}
The highest achieved score or count in the data set.
\item \textbf{Minimum}
The lowest achieved score or count in the data set.
\item \textbf{N}
The number of students evaluated.
\end{itemize}
\item \textbf{KR-21}
This statistic measures the internal reliability of the data set.
\textbf{This statistic is relevant for exams only.} Let $K$ be the
number of items in your test, $M$ the mean score on the test, $s$ the
standard deviation of the scores on your test, then the KR-21
statistic, $rk$, is given by the formula
$$ rk=\frac{K}{K-1}\cdot \frac{1-(M*(K-M))}{Ks^2} $$
\end{itemize}