DISCUSSION
Sampling distributions is a difficult topic for students to learn. A complete
understanding of sampling distributions requires students to integrate and apply
several concepts from different parts of a statistics course and to be able to reason
about the hypothetical behavior of many samples—a distinct, intangible thought
process for most students. The Central Limit Theorem provides a theoretical model
of the behavior of sampling distributions, but students often have difficulty mapping
this model to applied contexts. As a result, students fail to develop a deep
understanding of the concept of sampling distributions and therefore often develop
only a mechanical knowledge of statistical inference. Students may learn how to
compute confidence intervals and carry out tests of significance, but they are not
able to understand and explain related concepts, such as interpreting a p-value.
Most instructors have welcomed the introduction of simulation software and
web applets that allow students to visualize the abstract process of drawing repeated
random samples from different populations to construct sampling distributions.
However, our series of studies revealed that several ways of using such software
were not sufficient to affect meaningful change in students’ misconceptions of
sampling distributions. Despite the ability of a software program to offer interactive,
dynamic visualizations, students tend to look for rules and patterns and rarely
understand the underlying relationships that cause the patterns they see. For
example, students noticed that the sampling distribution became narrower and more
normal as the sample size increased, but did not necessarily understand why this was
happening. Therefore, when asked to make predictions about plausible distributions
of samples for a given sample size, students would resort to rules, often
misremembered or applied inconsistently, rather than think through the process that
might have generated these distributions. As a result, we often noticed students’
confusion when asked to distinguish between the distribution of one sample of data
and the distribution of several sample means.
By experimenting with different ways of having students interact with a
specially designed simulation program, we have explored ways to more effectively