...being just some introductory notes about research questions.

Research Questions in Single-Subject Research

This document presents information about how to determine whether a research question is appropriate for single-subject research methods and provides guidelines for writing questions in a form that promotes clarity in research design.

Just as is the case with other means of divining truth (prayer, casting coins, reading tea leaves, etc.), there are many ways to approach research questions.

Single-subject research methods examine the effects of two or more environmental conditions on the same individual. We compare the level of a person's behavior under one condition with the level of that same behavior under another condition. Thus, they are called "within-subject designs" because the same participant is observed in each condition.

A second important feature is that single-subject methods require experimental manipulation of the independent variable. The researcher changes the conditions to see what happens. Thus, for example, single-subject methods are not useful for examining the "normal" course of development, because aging is not a manipulable variable.

Examples of Questions

For examples, consider the following:

 Example Is SSD Suitable? Rationale 1. Whether left-brained students acquire more calculus skills than right-brained students. No The comparison is between types of people. 2. Whether left-brained students acquire more calculus skills when taught using Kline's Kind Kalculi program than when taught using Skiller's Drill program. Yes Although one might wonder about the type of people identified, the comparison is between the two types of instruction. 3. Whether people with dark hair are more likely to have dandruff than people with light hair. No The question is primarily about differences between people, not about whether a person's behavior differs under certain conditions 4. Whether a person's dandruff is greater when using Downflake® shampoo than when using Itchy-Itch Goo® shampoo. Yes The independent variable (type of shampoo) can be manipulated, and the results of that manipulation can be compared by examining the level of the dependent variable (dandruff) within an individual. 5. Whether 6-, 8-, or 10-year olds' accuracy in hitting a target with a thrown ball is greater when the distance of the target is 5, 10, or 15 meters away. No Although the question has different conditions (distance to the target), this part of the question refers to a performance that one would be unlikely to measure repeatedly over time. The first part of the question (about age) is both between-groups and not open to experimental manipulation. 6. Whether showing players a video of themselves performing a throwing task accurately just before their throwing performance is assessed affects their accuracy in hitting a target with a thrown ball when the distance of the target is 5, 10, or 15 meters away. Yes In this example, one can assess the performance at the different distances as separate dependent variables and manipulate the independent variable of showing the video. 7. Whether students' solution of novel algebra problems involving 2 unknowns is greater when they have been provided manipulatives than when they receive procedural instruction. Yes Although the question refers to more than one subject, the question rests on comparison of their performance under two different conditions. 8. Whether more verbally-facile students who receive procedural instruction have higher scores than more verbally-facile who receive manipulatives instruction but intuitively-facile students who receive manipulatives instruction have higher scores than intuitively-facile students who receive procedural instruction. No Although the question involves a manipulable independent variable (manipulative versus procedural instruction), it still rests on a difference between types of individuals (verbally- versus intuitively-facile types of students). [Note that I made up these types of students!]

Format for Question

Research questions sometimes don't take the grammatical form of the interrogative, but well-phrased questions can readily be converted to the interrogative form. Consider example 3 from the previous table. It could be restated in the interrogative form simply by deleting the word "whether" and moving the word "is" to the beginning of the sentence, as follows:

Is a person's dandruff greater when using Downflake® shampoo than when using Itchy-Itch Goo® shampoo.

There are many appropriate ways to phrase questions, and these are influenced by the question itself (e.g., whether it is correlational in nature) and on the methods used to investigate the question. Because single-subject methods are experimental in nature, the format of questions for single-subject studies should make clear that aspect of a single-subject study.

Research questions should reflect this experimental comparison. The researcher wants to know whether the level of the dependent variables is different under the different conditions. Questions composed of several specific parts readily show this relationship.

Further Information

A chapter by Lloyd, Tankersley, and Talbott (1994) provides more detailed treatment of this topic.

 SSD Home Announcements Assignments Calendar Lectures Resources References Curry