Sometimes, participants in an experiment can change the outcome.
In this lesson, we’ll look at some of the extraneous variables caused by participants: self-selection bias, demand characteristics, and good-subject bias.
Ollie is doing a study on the effects of light on factory workers. Specifically, he wants to know if brighter lights will result in more productive work. He goes to a factory that makes computer parts and asks for volunteers for a study.
After he gets his volunteers, he puts them into a room of their own and asks them to make computer parts as usual. Each week, he totals up the number of computer parts they make. Also each week, he adjusts the level of light they have in the room. If what he is thinking is right, then the weeks where the lights are brightest will produce the most computer parts.
Internal validity is the extent to which a study proves that only the independent variable is causing the changes in the dependent variable. In Ollie’s case, he wants to prove that the light level (the independent variable) is causing changes in the productivity of the workers (the dependent variable). But wait. What if the brighter lights make the room warmer? Because they feel hotter, the subjects might not do as much work. This is an example of an extraneous variable, or a factor besides the independent variable that affects the dependent variable. There are several types of extraneous factors.
Let’s look at a few that involve the participants.
Remember that Ollie asked for volunteers from among the workers at the factory. Psychological studies are done on volunteers, which poses a threat to internal validity known as self-selection bias. Think about this: what types of workers at the factory are most likely to volunteer for Ollie’s study? Probably workers who are enthusiastic.
They are most likely the same types of people who will volunteer to take on additional responsibility. They might be better at their jobs than non-volunteers, who just don’t care one way or another.Because they are enthusiastic go-getters, the subjects themselves might be the type of people who are always trying to improve and be more productive. As a result, Ollie won’t know whether the results are because of his manipulation of the lights or because the self-selected subjects are the type of people who are regularly trying to be more productive.
Another participant-influenced extraneous variable is known as demand characteristics, or the Hawthorne effect. This happens when subjects figure out what the researcher is studying and change their behavior accordingly. For example, what if the workers in Ollie’s study notice that the only thing changing from week to week is the level of light in the room where they’re working? They might suspect that Ollie is studying the effect of light on productivity.
But, what if they want the lights to be at a certain level? They might work to be more productive the week that the lights are at the level they prefer. In that way, they are changing the outcome with their behavior, and Ollie won’t ever know if it’s because of the Hawthorne effect or because the levels of light influence productivity.
Everyone wants to be liked, which is why the good-subject and social desirability biases come into play.
They are essentially the same thing: when good-subjects respond to an experimenter in order to be liked or seen as a good-subject. For example, what if the factory workers want Ollie’s approval? They don’t want him to be disappointed or dislike them, so they try to do their best, most productive work. But, are they being more productive because they want Ollie to like them or because the light levels are changing from week to week?The good-subject bias is most common in experiments that involve asking questions.
The subjects will try to answer in a way that will please the researcher. Even if Ollie just says something like, ‘Good job,’ at the end of the week, that could influence the next week’s work.
Internal validity is the extent to which an experimenter can prove that only the independent variable causes changes in the dependent variable. Extraneous variables, or factors other than the independent variable that can influence the dependent variable, can lower internal validity. Three major types of participant-based extraneous variables are self-selection bias, demand characteristics, which is also known as the Hawthorne effect, and good-subject and social desirability biases.
After reviewing this lesson, you should have the ability to:
- Define internal validity and extraneous variables
- Describe the major types of participant-based extraneous variables that can influence a psychological study