Introduction to Statistics

Goals:

1) Experience statistics as it is practiced by researchers in psychology:

Collect data appropriate to a specified purpose, and recognize limitations in existing data.
Explain the benefits of the statistical approach to design of experiments and use it.
Analyze data using appropriate graphs and numerical tools (primarily ANOVA).
Derive appropriate, actionable conclusions from data analysis.
Present results and conclusions in both technical and non-technical terms, in writing and orally.

2) Develop a systematic model for development of an experimental design:

Describe the role of statistical thinking and methods for problem solving.
Discuss the value of understanding, quantifying, and reducing variation.
Develop an understanding of how statistics is integral to research in cognitive psychology.
Determine the appropriate questions statistical consultants should ask.
Develop the ability to read scientific literature.

Materials:

Computer lab space is needed to conduct the experiment, which can be done inside or outside of the regularly scheduled class time.
In any experiment involving humans, Institutional Review Board approval should be received from your institution. It may be best to receive a general approval for all Simon experiments in your course. Even if you do receive general approval, it may be appropriate for the students to prepare an IRB proposal as a class assignment, even if it is not sent to the board.
Surprenant's article should be found found on-line at http://www.ingentaconnect.com/content/psocpubs/prp/2001/00000063/00000004;jsessionid=lqpqjsrrc3op.henrietta or other library resource such as http://www.apa.org/psycinfo/ for your class.
Assign each student (or group of students) one student ID and decide before class what course ID you plan on using. Any alphanumeric code without spaces is acceptable. However, human responses should be confidential and actual names should not be used as student IDs. You may want to search the database to determine if a course ID has already been used.
Other materials may be needed depending on student design.

On Line Simon Game:

Simon is an electronic game that was popular in the 1980s in which a person is expected to repeat a sequence of buttons provided by an electronic device. Each time you successfully repeat the given sequence of buttons, the sequence gets longer. The challenge is to remember as long a sequence as possible.

Cognitive psychologists test serial recall, which evaluates the ability of people to recall information in the specified order in which it was presented. Measuring how many items a subject can remember in order without an error, called memory span, is also studied. The Simon Game is an example of serial recall and memory span.

This on-line game provides students the opportunity to design multiple versions of the Simon Game in order to test which variables have the largest effect on memory. You can leave all the variables blank when you are simply trying out the game, however, if you want to find your score in the database of results, input any specific course ID and student ID.

The following link allows you to play the Simon Game.

In the following lab, students will have the opportunity to design and analyze an experiment to determine which factors are the most influential in successfully playing the Simon Game.

Simon Lab: Printable Student Version
Some older versions of Adobe Acrobat may have some trouble displaying the pdf files. You can download the most recent version of Adobe Acrobat Reader at www.adobe.com.

Instructors Notes:

Prerequisites: This lab can easily be done as a 2-sample t-test or a paired t-test using one explanatory variable. If students are planning to conduct a multifactor experiment, they should read and answer questions from the ANOVA: Factorial Designs tutorial before Day 2. More complex, within subjects designs are discussed in ANOVA: Advanced Designs.

Paper Review: I have found that some students are initially frustrated or intimidated when they are asked to read a research paper outside of their major. In my experience, providing more time to read the paper has not been helpful. It may, however, be appropriate to:

clarify that the paper review questions are worth only a small amount of the overall grade
allow students to complete the assignment in groups
allow students to turn in a revised version of their paper review questions after they are discussed in class

Class Discussion: I suggest spending at least part of three separate class periods to discuss the Surprenant (2001) paper and develop a class project or several group projects. It is helpful to initially identify that your goal as a professor is not to be able to answer all of their questions about current research in cognitive psychology, but to teach them a process in which they can find their own answers.

Since the Surprenant paper is somewhat complex, you may want to consider additional discussion topics:

You can instruct the students to go to the library and read the chapter on memory/remembering from any introductory psychology book. Ask them to identify some additional factors that influence memory.
http://www.muskingum.edu/~cal/database/general/memory.html is a nice site that discusses how to develop a better memory.
If you want students to find their own research papers on serial recall or memory span, http://www.apa.org/psycinfo/ is a good psychology resource. However, there are some difficulties with conducting a literature search for papers with tests similar to this Simon Game. Many papers have multiple tests with only a few containing sounds or tones. Since tone is typically not in the title of these papers, they are hard to find. In addition, tones are often the control conditions for words, so many papers primarily testing word memory will show up in the search.

Day 1: Spend at least 20 minutes discussing the paper review questions. Students may not feel qualified to design a psychology study, but they often can suggest appropriate modifications to a given study. Start students thinking about their own primary research questions and ask them:

How would you design a test to answer these questions?
Which factors and what levels could be used?
Which questions relate to main effects and which relate to interaction effects?
How can you design an experiment that is general enough to be of scientific interest, but specific enough that it can be conducted within a lab in a few weeks time?

Day 2: Invite a psychologist to the class (or a statistician if you are a psychologist) to answer student questions that are beyond your scope of knowledge.

It was very helpful for my students to hear from a researcher that he/she did not know all of the answers to their questions.
My students seemed to be somewhat surprised that they could contribute to science by conducting an experiment that has never been specifically studied before. Having a scientist in each field enforce the idea that science in not just a collection of facts but an ongoing exploratory process was very helpful for my students.

Day 3: After students have submitted their experimental ideas, the class can vote on one experiment to conduct as an entire class. This allows you to test more factors and levels with replicates in less time. However, each group of students could just as easily design and analyze their own experiments. Note that there are options for at least additional explanatory variables. My students having included time of day and whether or not a student ate breakfast in their experiments.

Final Paper Discussion: During the in-class review day, spend some time discussing how the process details (which are rarely discussed in textbooks and are often only given cursory comments in research papers) can significantly impact the analyses and conclusions of the data. My students are typically surprised by:

the amount of time and effort needed to appropriately conduct experiments
the numerous occasions in which an experiment can so easily become biased

Grading: I would suggest 50 points for the entire project (this is the same value as an exam in my course).

5 points for Day 1, the Surprenant review
10 points for Day 2, experimental design
5 points for lab procedures
5 points for appropriate comments and suggestions on other student papers
25 points for the final paper

If your institution does not have an Institutional Review Board, you can find more information on registration as well as educational materials at http://www.hhs.gov/ohrp/

Simon Game - Memory