What Is A Control Condition

Affiliate 6: Experimental Enquiry

Experimental Design

Explain the difference betwixt betwixt-subjects and within-subjects experiments, list some of the pros and cons of each approach, and decide which arroyo to employ to answer a particular enquiry question.
Ascertain random assignment, distinguish it from random sampling, explicate its purpose in experimental enquiry, and use some simple strategies to implement it.
Define what a command condition is, explain its purpose in inquiry on handling effectiveness, and describe some alternative types of command weather.
Define several types of carryover outcome, give examples of each, and explain how counterbalancing helps to deal with them.

In this section, we look at some different means to pattern an experiment. The master stardom we will make is between approaches in which each participant experiences one level of the independent variable and approaches in which each participant experiences all levels of the independent variable. The old are called between-subjects experiments and the latter are chosen within-subjects experiments.

Between-Subjects Experiments

In a , each participant is tested in only one condition. For example, a researcher with a sample of 100university students might assign one-half of them to write nearly a traumatic event and the other half write about a neutral event. Or a researcher with a sample of 60 people with severe agoraphobia (fear of open up spaces) might assign xx of them to receive each of three different treatments for that disorder. It is essential in a between-subjects experiment that the researcher assign participants to conditions and so that the different groups are, on average, highly similar to each other. Those in a trauma status and a neutral condition, for example, should include a similar proportion of men and women, and they should have similar average intelligence quotients (IQs), like average levels of motivation, similar average numbers of wellness problems, then on. This matching is a thing of controlling these extraneous participant variables across conditions so that they do non become confounding variables.

Random Assignment

The master way that researchers accomplish this kind of control of inapplicable variables beyond conditions is called , which ways using a random process to make up one's mind which participants are tested in which conditions. Do non confuse random assignment with random sampling. Random sampling is a method for selecting a sample from a population, and information technology is rarely used in psychological research. Random assignment is a method for assigning participants in a sample to the different atmospheric condition, and it is an of import element of all experimental enquiry in psychology and other fields too.

In its strictest sense, random assignment should encounter two criteria. Ane is that each participant has an equal chance of beingness assigned to each condition (eastward.g., a fifty% gamble of existence assigned to each of two weather). The second is that each participant is assigned to a condition independently of other participants. Thus one way to assign participants to two conditions would be to flip a coin for each one. If the money lands heads, the participant is assigned to Status A, and if it lands tails, the participant is assigned to Condition B. For three conditions, ane could utilize a computer to generate a random integer from ane to 3 for each participant. If the integer is 1, the participant is assigned to Condition A; if it is 2, the participant is assigned to Condition B; and if information technology is iii, the participant is assigned to Status C. In practice, a full sequence of conditions—one for each participant expected to exist in the experiment—is unremarkably created ahead of fourth dimension, and each new participant is assigned to the next status in the sequence as he or she is tested. When the procedure is computerized, the computer programme often handles the random assignment.

One trouble with coin flipping and other strict procedures for random assignment is that they are likely to result in unequal sample sizes in the unlike conditions. Unequal sample sizes are generally non a serious problem, and you should never throw away data you lot take already collected to achieve equal sample sizes. Yet, for a stock-still number of participants, it is statistically most efficient to divide them into equal-sized groups. It is standard do, therefore, to use a kind of modified random consignment that keeps the number of participants in each grouping as similar as possible. I approach is . In block randomization, all the conditions occur once in the sequence before any of them is repeated. Then they all occur again earlier whatsoever of them is repeated again. Inside each of these "blocks," the conditions occur in a random order. Once again, the sequence of conditions is normally generated before any participants are tested, and each new participant is assigned to the next status in the sequence. Table half-dozen.2 shows such a sequence for assigning nine participants to 3 atmospheric condition. The Research Randomizer website will generate block randomization sequences for any number of participants and conditions. Over again, when the procedure is computerized, the computer program often handles the block randomization.

Table half-dozen.iii Block Randomization Sequence for Assigning Nine Participants to Three Conditions
Participant	Condition
1	A
2	C
3	B
iv	B
five	C
6	A
7	C
viii	B
9	A

Random assignment is not guaranteed to control all extraneous variables across conditions. It is always possible that but by chance, the participants in one condition might turn out to be substantially older, less tired, more than motivated, or less depressed on average than the participants in another condition. Yet, there are some reasons that this possibility is not a major business organization. One is that random assignment works amend than one might expect, especially for big samples. Another is that the inferential statistics that researchers utilize to decide whether a difference betwixt groups reflects a difference in the population takes the "fallibility" of random assignment into account. Withal some other reason is that even if random assignment does event in a confounding variable and therefore produces misleading results, this confound is probable to be detected when the experiment is replicated. The consequence is that random assignment to conditions—although not infallible in terms of decision-making extraneous variables—is always considered a strength of a inquiry design.

Treatment and Command Weather

Betwixt-subjects experiments are often used to decide whether a treatment works. In psychological research, a is any intervention meant to alter people'southward behaviour for the better. This intervention includes psychotherapies and medical treatments for psychological disorders but also interventions designed to improve learning, promote conservation, reduce prejudice, and then on. To determine whether a treatment works, participants are randomly assigned to either a , in which they receive the handling, or a , in which they do not receive the treatment. If participants in the treatment status cease up better off than participants in the control condition—for example, they are less depressed, learn faster, conserve more, express less prejudice—then the researcher can conclude that the treatment works. In research on the effectiveness of psychotherapies and medical treatments, this type of experiment is often called a .

There are dissimilar types of control weather condition. In a , participants receive no treatment whatsoever. One problem with this approach, however, is the existence of placebo effects. A is a fake treatment that lacks any active ingredient or element that should make it effective, and a is a positive issue of such a treatment. Many folk remedies that seem to work—such as eating craven soup for a cold or placing soap nether the bedsheets to stop nighttime leg cramps—are probably nothing more than than placebos. Although placebo effects are non well understood, they are probably driven primarily by people'southward expectations that they volition improve. Having the expectation to ameliorate tin can result in reduced stress, anxiety, and depression, which tin modify perceptions and even better immune system functioning (Toll, Finniss, & Benedetti, 2008) ^[ane].

Placebo effects are interesting in their own right (meet Note "The Powerful Placebo" ), simply they likewise pose a serious problem for researchers who want to determine whether a handling works. Figure 6.2 shows some hypothetical results in which participants in a treatment condition improved more than on boilerplate than participants in a no-treatment control condition. If these atmospheric condition (the two leftmost bars in Figure half dozen.2 ) were the simply atmospheric condition in this experiment, yet, one could not conclude that the treatment worked. Information technology could be instead that participants in the treatment grouping improved more because they expected to improve, while those in the no-treatment command condition did non.

Figure half-dozen.2 Hypothetical Results From a Written report Including Handling, No-Treatment, and Placebo Weather

Fortunately, in that location are several solutions to this problem. Ane is to include a , in which participants receive a placebo that looks much like the treatment merely lacks the active ingredient or element idea to be responsible for the handling'due south effectiveness. When participants in a treatment condition take a pill, for case, and then those in a placebo control status would accept an identical-looking pill that lacks the active ingredient in the treatment (a "sugar pill"). In research on psychotherapy effectiveness, the placebo might involve going to a psychotherapist and talking in an unstructured style near 1'south problems. The thought is that if participants in both the handling and the placebo control groups wait to better, so any comeback in the treatment group over and above that in the placebo command group must have been caused past the treatment and not by participants' expectations. This divergence is what is shown by a comparison of the ii outer bars in Figure six.2 .

Of course, the principle of informed consent requires that participants be told that they will be assigned to either a treatment or a placebo command condition—even though they cannot exist told which until the experiment ends. In many cases the participants who had been in the control condition are so offered an opportunity to take the real treatment. An culling approach is to use a , in which participants are told that they volition receive the handling merely must await until the participants in the treatment status have already received it. Thisdisclosureallows researchers to compare participants who have received the treatment with participants who are not currently receiving it but who still expect to improve (eventually). A final solution to the problem of placebo effects is to leave out the control condition completely and compare any new treatment with the best available culling treatment. For example, a new treatment for simple phobia could be compared with standard exposure therapy. Because participants in both conditions receive a treatment, their expectations well-nigh improvement should be similar. This approach also makes sense considering once there is an effective treatment, the interesting question about a new handling is non simply "Does it work?" only "Does it piece of work better than what is already available?

Many people are non surprised that placebos tin have a positive effect on disorders that seem fundamentally psychological, including depression, anxiety, and insomnia. Notwithstanding, placebos can also accept a positive issue on disorders that most people retrieve of as fundamentally physiological. These include asthma, ulcers, and warts (Shapiro & Shapiro, 1999) ^[2]. There is even prove that placebo surgery—also chosen "sham surgery"—can be as constructive as actual surgery.

Medical researcher J. Bruce Moseley and his colleagues conducted a study on the effectiveness of 2 arthroscopic surgery procedures for osteoarthritis of the human knee (Moseley et al., 2002) ^[iii].The control participants in this study were prepped for surgery, received a tranquilizer, and fifty-fifty received iii modest incisions in their knees. Merely they did non receive the actual arthroscopic surgical process. The surprising result was that all participants improved in terms of both knee hurting and function, and the sham surgery group improved merely as much as the treatment groups. According to the researchers, "This study provides strong show that arthroscopic lavage with or without débridement [the surgical procedures used] is not ameliorate than and appears to be equivalent to a placebo procedure in improving human knee pain and self-reported function" (p. 85).

Within-Subjects Experiments

In a , each participant is tested under all conditions. Consider an experiment on the effect of a defendant's physical attractiveness on judgments of his guilt. Again, in a between-subjects experiment, 1 group of participants would be shown an attractive accused and asked to judge his guilt, and another grouping of participants would be shown an unattractive defendant and asked to guess his guilt. In a inside-subjects experiment, however, the same grouping of participants would estimate the guilt of both an bonny and an unattractive accused.

The principal reward of this arroyo is that information technology provides maximum command of extraneous participant variables. Participants in all conditions have the same mean IQ, same socioeconomic condition, aforementioned number of siblings, and so on—considering they are the very same people. Within-subjects experiments also make information technology possible to use statistical procedures that remove the effect of these extraneous participant variables on the dependent variable and therefore make the data less "noisy" and the effect of the independent variable easier to detect. We will look more closely at this idea afterwards in the book. Nonetheless, non all experiments can employ a within-subjects pattern nor would it be desirable to.

Carryover Effects and Counterbalancing

The primary disadvantage of within-subjects designs is that they tin result in carryover furnishings. A is an result of beingness tested in i status on participants' behaviour in afterward conditions. I type of carryover effect is a , where participants perform a job better in afterwards weather considering they have had a chance to practice information technology. Another type is a , where participants perform a task worse in later conditions because they get tired or bored. Being tested in ane status tin also change how participants perceive stimuli or interpret their task in later on conditions. This type of effect is called a . For case, an average-looking defendant might be judged more harshly when participants have just judged an attractive defendant than when they have just judged an unattractive defendant. Within-subjects experiments as well make it easier for participants to guess the hypothesis. For example, a participant who is asked to judge the guilt of an attractive defendant and then is asked to judge the guilt of an unattractive accused is probable to guess that the hypothesis is that accused attractiveness affects judgments of guilt. This knowledge could pb the participant to guess the unattractive defendant more harshly because he thinks this is what he is expected to do. Or it could make participants judge the 2 defendants similarly in an try to be "fair."

Carryover furnishings tin can be interesting in their ain correct. (Does the bewitchery of one person depend on the attractiveness of other people that we have seen recently?) Only when they are non the focus of the research, carryover effects tin can be problematic. Imagine, for example, that participants judge the guilt of an attractive defendant and then judge the guilt of an unattractive accused. If they judge the unattractive accused more harshly, this might be because of his unattractiveness. Merely it could be instead that they approximate him more harshly because they are becoming bored or tired. In other words, the order of the conditions is a confounding variable. The attractive condition is always the first condition and the unattractive condition the 2d. Thus any departure between the weather condition in terms of the dependent variable could be caused by the order of the atmospheric condition and not the independent variable itself.

There is a solution to the trouble of order furnishings, however, that can be used in many situations. Information technology is, which means testing different participants in dissimilar orders. For instance, some participants would be tested in the bonny defendant status followed past the unattractive defendant condition, and others would be tested in the unattractive condition followed by the bonny condition. With three conditions, there would exist six different orders (ABC, ACB, BAC, BCA, CAB, and CBA), so some participants would be tested in each of the half-dozen orders. With counterbalancing, participants are assigned to orders randomly, using the techniques we accept already discussed. Thus random assignment plays an of import role in inside-subjects designs but as in betwixt-subjects designs. Here, instead of randomly assigning to conditions, they are randomly assigned to dissimilar orders of conditions. In fact, it can safely be said that if a study does not involve random assignment in one form or another, information technology is not an experiment.

An efficient way of counterbalancing is through a Latin square pattern which randomizes through having equal rows and columns. For example, if you have iv treatments, you must have four versions. Like a Sudoku puzzle, no treatment can repeat in a row or column. For four versions of four treatments, the Latin square design would await similar:

A	B	C	D
B	C	D	A
C	D	A	B
D	A	B	C

There are two ways to call back about what counterbalancing accomplishes. I is that it controls the order of conditions and then that it is no longer a misreckoning variable. Instead of the attractive condition always being first and the unattractive status always beingness second, the attractive status comes get-go for some participants and second for others. Likewise, the unattractive condition comes first for some participants and 2nd for others. Thus whatever overall difference in the dependent variable betwixt the two weather cannot take been caused by the lodge of weather. A 2nd mode to think about what counterbalancing accomplishes is that if at that place are carryover effects, it makes it possible to observe them. I can analyze the data separately for each order to meet whether it had an outcome.

Researcher Michael Birnbaum has argued that the lack of context provided by between-subjects designs is often a bigger problem than the context effects created by inside-subjects designs. To demonstrate this problem, he asked participants to charge per unit ii numbers on how large they were on a calibration of 1-to-10 where ane was "very very small" and 10 was "very very large". One grouping of participants were asked to rate the number nine and another group was asked to rate the number 221 (Birnbaum, 1999)^[iv]. Participants in this between-subjects blueprint gave the number 9 a mean rating of v.13 and the number 221 a mean rating of three.ten. In other words, they rated nine as larger than 221! According to Birnbaum, this deviation is because participants spontaneously compared ix with other i-digit numbers (in which example it is relatively large) and compared 221 with other iii-digit numbers (in which case it is relatively small-scale).

Simultaneous Within-Subjects Designs

Then far, nosotros have discussed an approach to inside-subjects designs in which participants are tested in one status at a time. There is another approach, however, that is oft used when participants make multiple responses in each condition. Imagine, for example, that participants judge the guilt of ten bonny defendants and x unattractive defendants. Instead of having people make judgments about all x defendants of 1 type followed by all 10 defendants of the other type, the researcher could present all twenty defendants in a sequence that mixed the 2 types. The researcher could then compute each participant's hateful rating for each type of accused. Or imagine an experiment designed to see whether people with social anxiety disorder remember negative adjectives (e.g., "stupid," "incompetent") ameliorate than positive ones (east.yard., "happy," "productive"). The researcher could have participants written report a unmarried list that includes both kinds of words and so accept them endeavor to recall as many words every bit possible. The researcher could then count the number of each blazon of discussion that was recalled. There are many ways to decide the order in which the stimuli are presented, simply one mutual style is to generate a different random gild for each participant.

Between-Subjects or Inside-Subjects?

Nigh every experiment can be conducted using either a between-subjects pattern or a within-subjects design. This possibility means that researchers must choose betwixt the two approaches based on their relative merits for the particular situation.

Between-subjects experiments have the advantage of being conceptually simpler and requiring less testing time per participant. They too avoid carryover effects without the demand for counterbalancing. Within-subjects experiments have the advantage of controlling extraneous participant variables, which more often than not reduces racket in the data and makes information technology easier to detect a relationship between the independent and dependent variables.

A skillful dominion of thumb, then, is that if it is possible to conduct a within-subjects experiment (with proper counterbalancing) in the fourth dimension that is available per participant—and you lot take no serious concerns about carryover effects—this design is probably the best option. If a within-subjects design would be difficult or impossible to deport out, and so you should consider a betwixt-subjects design instead. For example, if you were testing participants in a medico'south waiting room or shoppers in line at a grocery store, you might not have enough fourth dimension to test each participant in all conditions and therefore would opt for a between-subjects pattern. Or imagine you were trying to reduce people'southward level of prejudice by having them interact with someone of another race. A within-subjects design with counterbalancing would require testing some participants in the handling status kickoff and so in a command condition. But if the treatment works and reduces people'southward level of prejudice, then they would no longer be suitable for testing in the control condition. This difficulty is true for many designs that involve a treatment meant to produce long-term modify in participants' behaviour (e.g., studies testing the effectiveness of psychotherapy). Clearly, a betwixt-subjects design would exist necessary hither.

Retrieve also that using one blazon of pattern does not preclude using the other type in a unlike study. There is no reason that a researcher could non use both a between-subjects design and a within-subjects design to answer the same research question. In fact, professional researchers oftentimes take exactly this type of mixed methods approach.

Experiments can be conducted using either between-subjects or within-subjects designs. Deciding which to use in a particular situation requires careful consideration of the pros and cons of each approach.
Random assignment to conditions in between-subjects experiments or to orders of weather condition in inside-subjects experiments is a primal chemical element of experimental research. Its purpose is to control extraneous variables so that they do not become confounding variables.
Experimental research on the effectiveness of a treatment requires both a treatment condition and a command condition, which can be a no-treatment control condition, a placebo control condition, or a waitlist control condition. Experimental treatments tin also be compared with the best available culling.

Discussion: For each of the following topics, listing the pros and cons of a between-subjects and within-subjects blueprint and decide which would be better.
1. You desire to test the relative effectiveness of two grooming programs for running a marathon.
2. Using photographs of people as stimuli, you desire to see if smiling people are perceived equally more intelligent than people who are not smiling.
3. In a field experiment, you want to encounter if the manner a panhandler is dressed (neatly vs. sloppily) affects whether or not passersby give him any money.
4. You desire to see if concrete nouns (e.grand., dog ) are recalled better than abstract nouns (due east.g., truth ).
Discussion: Imagine that an experiment shows that participants who receive psychodynamic therapy for a canis familiaris phobia improve more than participants in a no-handling command group. Explicate a primal trouble with this research blueprint and at least two means that it might exist corrected.