Implementation tips

General comments on the Project Ideas

The list of research questions is a critical stepping stone to a successful project. Without your review of this list as a preliminary filter, the students might put a lot of time and energy into a research proposal that is inadequate. By reviewing their list of research questions, you will be helping them to select an idea that will provide the right learning experience for the students.

In my syllabus, the list of project ideas is a project requirement, but does not count for any points toward the students’ final grades. This is intended to reduce the amount of stress associated with this requirement so that the students feel freer to brainstorm and generate creative ideas.

Common things to avoid in a list of research questions

When reviewing a list of research questions, there are common situations you should steer the students away from.

Try to avoid the following types of research questions:

Done in a previous semester at your college
Common Internet examples
Questions with relatively obvious answers

Try to avoid research questions with the following types of data:

Readily available data
Data collection doesn’t involve random sampling
Data collection where only one person is involved

Try to correct these common errors when they come up in the research questions:

Ambiguous population
Ambiguous variables
Confusing the three types of questions: describing, comparing or relating

Describing, comparing, or relating?

Not surprising, students are not used to thinking of their world in terms of statistical analysis techniques. Accordingly, the issue of whether a research question is describing, comparing, or relating can be very difficult.

There are many ways to approach this question to make it accessible to the students. The method that works best for me is used subtly in the textbook. Among the bullets in section 3.2 are “Ask a question about a large group” and “Consider what will be recorded.” The first of these is the population. The second of these is the variable (or variables). If the students are having trouble decided whether their research questions are describing, comparing or relating, walk them through these two questions first.

For example, if a group wants to submit “Number of seeds in a watermelon compared to the size of the watermelon,” but doesn’t know whether this is describing, comparing, or relating, make sure that they first understand that the population consists of watermelons. Next make sure they understand that they will need to record the number of seeds in a watermelon and the weight (or possibly diameter) of the watermelon (for size).

With this information, we are ready to tackle the main question. When we state our conclusion, will we say simply how many seeds and how big the watermelons are (describing), that the watermelons have more seeds than they do weight (comparing) or will we say that the bigger the watermelon, the more seeds it has (relating)? Note that to an experienced researcher, the second question is meaningless because the units don’t match, but students often struggle with these concepts.

Why not allow opinion surveys?

The primary reason I do not allow opinion surveys for projects in my class is that the issues involved with opinion surveys are bigger than the objectives I am trying to accomplish with these projects. In my opinion (there’s that word again!), the students should learn to carry out objective research before trying to carry out subjective research.

Other reasons for not allowing opinion surveys is that the responses are often harder to quantify, harder to analyze and harder to interpret. While opinion surveys might be more interesting to the students than objective research, the effort required to do them right is beyond the level of most introductory statistics courses.

2004 McGraw-Hill Higher Education
Any use is subject to the Terms of Use and Privacy Policy.
McGraw-Hill Higher Education is one of the many fine businesses of The McGraw-Hill Companies.