How to do your own experiments in psychology and education
1c. Choose your factors
A wide range of things can affect how a psychological process works. You will study one and your partner will study another; together you'll check whether the factors affect each other.
Pick factors that you think will affect your sub-process directly. You will probably find published research on how different factors affect your sub-process. You can either do another study on a factor that other researchers have studied, or you can try a new factor.
Read this topic next: Formulate your research problem.
Examples with one factor:
The factors are different things can make your sub-process more error prone, slower, faster, more variable, etc. There may be several of these factors. You will only study two, but it is good to identify others.
Empirical research focuses on what causes what, so your problem will zoom in on a process, action, or some characteristic, not an object. If you focus on an object, like the heart, then your research will often focus on describing the parts of the object, how the are related, and perhaps how they change over time. If you focus on a process, like listening or deciding, then your research will focus on identifying the factors that lead to changes in the process, for example: noise with the stimuli, sleepiness of the participant, stress, etc.
The main question is: What actually determines how a process happens? To make your research problem more specific, you have to decide what you think will lead changes in your process. For example, does letter size affect word recognition? Does word length affect word recognition? Letter Size and Word Length here are examples of factors (which are measured by independent variables) – characteristics that the researcher thinks might cause a change in the sub-process studied (here, word recognition). By convention, we write factor names with capital letters, so that they stand out.
You cannot call this thing that might affect your sub-process a “cause” because you do not know yet if it really does cause anything and you do not want to prejudge the results. Most researchers will call it a “factor” because it may be a factor in the changes of the sub-process.
What kinds of things determine or cause changes in the sub-process that you want to study? Here are the usual kinds of factors that researchers investigate.
1. Do you think differences in people lead to differences in your sub-process? Maybe you think that Native language (Chinese speakers vs. Russian speakers) determines how people spell things. Maybe you think that Gender (men vs. women) determines how often people interrupt each other. Maybe you think that Expertise or Knowledge (Students vs. Experts) determines how you understand a particular kind of text. The effects of Age will be difficult for us to investigate – college students don’t vary very much in terms of age.
2. Do you think that differences in the materials or stimuli that you give them leads to differences in the sub-process you want to study? Maybe you think that Letter Size is important: smaller letters (e.g., 10 pt) are harder to read than larger letters (e.g., 18 pt) or that low-frequency words are harder to read than high-frequency words.
3. Maybe you think that doing different tasks with the same materials can lead to different things happening during your sub-process: maybe word recognition during reading for fun is different from word recognition during reading to answer questions. For example, it could be that writing to a future boss and to a future spouse will make you choose different words or different topics (I hope so!).
4. Do you think that the situation in which people do things affects what happens? Is reading a technical text in a silent room different from reading it while there is loud music or an interesting movie? Will a strong smell of popcorn affect how well children read?
Specifying the factor(s) that you want to study is the key step to focus your research problem. However, there is another reason to include the factors: they help the researcher keep on track when trying to relate the hypotheses (like the role of the factor) and observations (which we will collect later). The factors provide a link to the hidden mechanisms that have to be imagined or hypothesized. When you are focus on looking for hidden causes, it guarantees that you will reason about observations, rather than just collect them. Jacob, once again, summarizes this principle very clearly:
"In some respects, at least, myths and science fulfill a similar function: they both provide human beings with a representation of the world and of the forces that are supposed to govern it. They both fix the limits of what is considered as possible. (...) In their attempt to perform their function and to transform the chaos of the world into order, myths and scientific theories operate on the same principle. The [goal] is always to explain visible events by invisible forces, to connect what is seen with what is assumed. (...) A phenomenon is considered to be explained if it can be viewed as the visible effect of some hidden cause related to the whole network of invisible forces that are supposed to govern the world.” (Jacob, 1982, p. 359-362, emphasis added)
This is particularly important for formulating research problems: scientific problems include both a mechanism and the factors that can affect that mechanism. This is one reason why so many research articles have titles like “Effects of marijuana on short-term memory for numbers”: the researcher, in this case, is investigating how marijuana causes changes to the (invisible) mechanisms of short-term memory, which in turn cause observable effects in behavior that the researcher measured.
To judge the possible effects of a factor (like Letter Size in the example), researchers will design different experimental situations where everything is the same except for the factor: in the simplest situation, there would be two different levels of the factor. For example, an experiment might compare 10 pt letters with 18 pt letters. To keep the factor names clear, they are capitalized, so researchers talk about “Two levels (10 pt, 18 pt) of Letter Size”. The term “levels”, then, just refers to the different values that a given factor has in a particular experiment.
You have enough information to formulate the research problem in a much clearer way: as a simple yes/no question.
Now, then, the formulation of your example research problem is:
Do Letter Size and Word Frequency affect word recognition during reading?
The general format of the research problem, so far, is:
Do [factor and factor] affect [sub-process] during [process]?
Pick factors that you think will affect your sub-process directly. You will probably find published research on how different factors affect your sub-process. You can either do another study on a factor that other researchers have studied, or you can try a new factor.
Read this topic next: Formulate your research problem.
Examples with one factor:
- how does the presence of a picture [factor] affect how people use topic knowledge [sub-process] to comprehend [process] a story?
- how does the weight of a person [factor] affect how other people form stereotypes [sub-process] when they see [process] him or her?
- how does sentence length [factor] affect how well people put together the parts of the sentence [sub-process] when they're listening [process]?
- how does the presence of music [factor] affect how well people multiply numbers [sub-process] in working memory [process]?
The factors are different things can make your sub-process more error prone, slower, faster, more variable, etc. There may be several of these factors. You will only study two, but it is good to identify others.
Empirical research focuses on what causes what, so your problem will zoom in on a process, action, or some characteristic, not an object. If you focus on an object, like the heart, then your research will often focus on describing the parts of the object, how the are related, and perhaps how they change over time. If you focus on a process, like listening or deciding, then your research will focus on identifying the factors that lead to changes in the process, for example: noise with the stimuli, sleepiness of the participant, stress, etc.
The main question is: What actually determines how a process happens? To make your research problem more specific, you have to decide what you think will lead changes in your process. For example, does letter size affect word recognition? Does word length affect word recognition? Letter Size and Word Length here are examples of factors (which are measured by independent variables) – characteristics that the researcher thinks might cause a change in the sub-process studied (here, word recognition). By convention, we write factor names with capital letters, so that they stand out.
You cannot call this thing that might affect your sub-process a “cause” because you do not know yet if it really does cause anything and you do not want to prejudge the results. Most researchers will call it a “factor” because it may be a factor in the changes of the sub-process.
What kinds of things determine or cause changes in the sub-process that you want to study? Here are the usual kinds of factors that researchers investigate.
1. Do you think differences in people lead to differences in your sub-process? Maybe you think that Native language (Chinese speakers vs. Russian speakers) determines how people spell things. Maybe you think that Gender (men vs. women) determines how often people interrupt each other. Maybe you think that Expertise or Knowledge (Students vs. Experts) determines how you understand a particular kind of text. The effects of Age will be difficult for us to investigate – college students don’t vary very much in terms of age.
2. Do you think that differences in the materials or stimuli that you give them leads to differences in the sub-process you want to study? Maybe you think that Letter Size is important: smaller letters (e.g., 10 pt) are harder to read than larger letters (e.g., 18 pt) or that low-frequency words are harder to read than high-frequency words.
3. Maybe you think that doing different tasks with the same materials can lead to different things happening during your sub-process: maybe word recognition during reading for fun is different from word recognition during reading to answer questions. For example, it could be that writing to a future boss and to a future spouse will make you choose different words or different topics (I hope so!).
4. Do you think that the situation in which people do things affects what happens? Is reading a technical text in a silent room different from reading it while there is loud music or an interesting movie? Will a strong smell of popcorn affect how well children read?
Specifying the factor(s) that you want to study is the key step to focus your research problem. However, there is another reason to include the factors: they help the researcher keep on track when trying to relate the hypotheses (like the role of the factor) and observations (which we will collect later). The factors provide a link to the hidden mechanisms that have to be imagined or hypothesized. When you are focus on looking for hidden causes, it guarantees that you will reason about observations, rather than just collect them. Jacob, once again, summarizes this principle very clearly:
"In some respects, at least, myths and science fulfill a similar function: they both provide human beings with a representation of the world and of the forces that are supposed to govern it. They both fix the limits of what is considered as possible. (...) In their attempt to perform their function and to transform the chaos of the world into order, myths and scientific theories operate on the same principle. The [goal] is always to explain visible events by invisible forces, to connect what is seen with what is assumed. (...) A phenomenon is considered to be explained if it can be viewed as the visible effect of some hidden cause related to the whole network of invisible forces that are supposed to govern the world.” (Jacob, 1982, p. 359-362, emphasis added)
This is particularly important for formulating research problems: scientific problems include both a mechanism and the factors that can affect that mechanism. This is one reason why so many research articles have titles like “Effects of marijuana on short-term memory for numbers”: the researcher, in this case, is investigating how marijuana causes changes to the (invisible) mechanisms of short-term memory, which in turn cause observable effects in behavior that the researcher measured.
To judge the possible effects of a factor (like Letter Size in the example), researchers will design different experimental situations where everything is the same except for the factor: in the simplest situation, there would be two different levels of the factor. For example, an experiment might compare 10 pt letters with 18 pt letters. To keep the factor names clear, they are capitalized, so researchers talk about “Two levels (10 pt, 18 pt) of Letter Size”. The term “levels”, then, just refers to the different values that a given factor has in a particular experiment.
You have enough information to formulate the research problem in a much clearer way: as a simple yes/no question.
Now, then, the formulation of your example research problem is:
Do Letter Size and Word Frequency affect word recognition during reading?
The general format of the research problem, so far, is:
Do [factor and factor] affect [sub-process] during [process]?
About
© 2009 Created by Mike Dillinger, PhD on Ning. Create a Ning Network!
