Confounding Exercise - Research Methods (Psychology)

8/13/2019 Confounding Exercise - Research Methods (Psychology)

1/17


2/17

1. The problem of whether psychological Statistics studentsshould be taught statistics in the computer lab usingsoftware or in a classroom using calculator-based analyseshas been a point of controversy for many years. To attemptto decide this issue anDr. Frank N. Furter plans to teachstatistics to two groups, one by each method. PBA teaches

only statistics using calculator-based math. "This is fine forone group," the Dr. Furter says. "Now I must find a collegethat used computer-based approach." Accordingly a visit ismade to another school that uses the computer-basedapproach, where a sample of students is tested to see howwell they can deal with statistical analysis and interpretation.After administering an extensive test, it is found that thestudents who learned using calculator-based course workare reliably superior to those who learned by the computer-based approach. It is then concluded that the calculator-based course is superior to the computer-based method. Doaccept or reject this conclusion? Why?


3/17

The confounding in this study is especiallyatrocious. The participants in the two groupsundoubtedly differ in a large number of respectsother than type of method. For instance, there may

be differences in intelligence, opportunity to study,socioeconomic level, as well as differences in mathand analytical proficiency prior to learning by eitherapproach, and certainly there were differentinstructors. The proper approach would be to

randomly assign participants from the same class ina given school to two groups, and then to randomlydetermine which group is taught by each method,both groups being taught by the same instructor.


4/17

2. An Industrial/Organizational Psychologist is interested inwhether training on a simulator facilitates accuracy in an actualassembly process. A group of new employees with noprevious assembly experience is randomly divided into twogroups. One group receives training on the simulator; theother does not. Both groups are then tested on their ability to

assemble the target commodity under actual assembly lineconditions. There are two assembly lines and associatedequipment. The simulator-trained group is assigned to oneassembly line and a corresponding set of target items, whereasthe control group assembles the target items on the secondassembly line. The tests show that the group previouslytrained on the simulator is significantly faster and make fewererrors than is the control group. The conclusion is thatsimulator training facilitates actual assembly line performance.


5/17

The characteristics of the individual assembly linesand target items are confounded with theindependent variable. It may be that the machineryis more accurate on one assembly line than the

other, and that one set of targets is easier to producethan the other. To control these variables one mighthave all participants assemble from the sameassembly line with the same run of target items. Orhalf of the participants from each group could

assemble from each assembly line with each set oftargets. The simulator itself gave that group extrapractice too, another confound.


6/17

3. A psychologist tests the hypothesis that early toilet trainingleads to a personality of excessive compulsiveness aboutcleanliness, and conversely, that late toilet training leads tosloppiness. Previous studies have shown that middle-classchildren receive their toilet training earlier than do lower-classchildren so that one group is formed of middle-class children and

another of lower-class children. Both groups are provided with afinger painting task, and such data are recorded as the extent towhich children smear their hands and arms with paints, whetherthey clean up after the session, and how many times they washthe paints from their hands. Comparisons of the two groups othese criteria indicate that the middle-class children are reliablymore concerned about cleanliness than are those of the lower-class. It is thus concluded that early toilet training leads tocompulsive cleanliness, whereas later toilet training results inless concern about personal cleanliness.


7/17

Undoubtedly these classes differ in a numberof respects, among which is age at whichthey are toilet trained. The dependentvariable results may thus be due to some

other difference between the groups such asamount of social stimulation or amount ofmoney spent on family needs. The obvious,but difficult, way to conduct this experimentin order to establish a causal relation would

be to randomly select a group of children,randomly assign them to two groups, andthen randomly determine the age at whicheach group is toilet trained.


8/17

4. A hypothesis is that emotionally loaded words like sexandp rost i tute must be exposed for a longer time to beperceived than neutral words, an idea important forconstruction of ads and other messages. To test this

hypothesis, various words are exposed to participantsfor extremely short intervals. In fact, the initialexposure time is so short that no participant canreport any of the words. The length of exposure isthen gradually increased until each word is correctly

reported. The length of exposure necessary for eachword to be reported is recorded. It is found that thelength of time necessary to report the emotionallyloaded words is longer than that for the neutral word.It is concluded that the hypothesis is confirmed.


9/17

There may be other reasons for not reporting anemotionally loaded word than that it is not perceived.For instance, sexmay actually be perceived, but theparticipant waits until being absolutely sure that thatis the word, possibly saving the person from a

"social blunder." In addition, the frequency withwhich the loaded and neutral words are used ineveryday life undoubtedly differs, thus affecting thethreshold for recognition of the words. A betterapproach would be to start with a number of words

that are emotionally neutral (or with nonsensesyllables), and make some of them emotionallyloaded (such as associating an electric shock orother negative consequence with them). The loadedand neutral words should be equated for frequencyof use.


10/17

5. A health psychologist conducted anexperiment to study the effect ofacupuncture on pain. Half of the participantswere treated for painful shoulders through

acupuncture, whereas the other half receivedno special treatment. The participants whoreceived acupuncture treatment reported areliable improvement in shoulder discomfortto a "blind" evaluator after treatment.

However, no statistically reliableimprovement was reported by the controlgroup. The physician concluded thatacupuncture is an effective treatment forchronic shoulder pain.


11/17

One should not accept this conclusion, because thereis no control for the effects of suggestion. Thecontrol group should have experienced sometreatment similar to that of the experimental group,

such as having a different pattern of needles insertedbeneath the skin. Studies that have controlled for theeffects of suggestion have indicated no suchdifferences between acupuncture and placebogroups. Other scientific studies independently have

confirmed that merely suggesting that pain will bereduced through experimental techniques is sufficientto lead patients to report decreased pain. Finally, itwould have been stronger to show that theexperimental group was superior to the control group.


12/17

6. Dr. Statsdud wants to study the effects of grades asrewards or punishments. He chooses twoIntroductory Psychology classes. Dr. Psycho teachesboth classes (i.e., both sections). In one classstudents were given A, B, C, D, or F grades, whereas

the other class either passed or failed. Testsindicated that there were no reliable differencesbetween the two classes in terms of achievement,attitudes, or values. The conclusion was that studentslearn just as well without the reward or punishment of

grades. Dr. Statsdud also observed a difference inclassroom atmosphere in which the pass-fail classwas more relaxed and free of grade-oriented tensionswith better rapport between the Dr. Psycho andstudents.


13/17

This research does not allow us to draw any conclusions because of its faultymethodology in a number of respects. The two groups of students probablydiffered before the research was started. Not randomly assigning students to thetwo classes, but allowing them to be selected on the basis of class hours,produces a confound. The instructor's belief in the relative efficacy of themethods may have influenced performance of students in the two classes. Hencethe students being taught by the pass-fail method may have outperformed whatwould have been normal for them. The most serious methodological criticism isthat failure to find a difference between groups does not allow one to concludethat the two methods are equally effective. There are an infinite number ofpossible reasons why two conditions in any study may not differ significantly,only one of which is that the (population) means on the dependent variablescores of the two groups are equal. Failure to reject the null hypothesis is notequivalent to accepting the null hypothesis. A much more likely reason for failingto find a reliable difference between groups is that there is excessiveexperimental error in the conduct of the research; typically this is due to poor

control methodology. Finally, casual observation of a difference in "classroomatmosphere" hardly provides the kind of information upon which educationalcurricula should be based. Regardless, as an extension of this kind ofconclusion, one could probably predict that a course in which there were nogrades at all would result in totalfreedom from "grade oriented tensions."


14/17

7. Wexley and Thornton (1972) conducted an experiment to test thehypothesis that providing employees in training with verbalfeedback about assessment results would facilitateimprovements. During the training period, 169 trainees weregiven four assessments. After each assessment, of theparticipants were reshown 18 of the 35 multiple-choice items.

The training instructor read each of these 18 items, and gavethe correct answer and a brief explanation of the correct answer.The instructor indicated that time did not permit going over all 35items; therefore participants did not see or receive feedback on17 items per quiz. The other of the trainees, the control group,were not reshown the items and, thus, did not receive verbalfeedback. The final assessment in the training period consistedof 38 of the feedback questions and 38 of the nonfeedbackquestions. Wexley and Thornton found that, in general, traineesdid better on items for which feedback was given than those forwhich feedback was not given. Wexley and Thornton concludedthat the results confirm the assumption that post-assessmentverbal feedback does facilitate workplace improvement.


15/17

The confounding variable in problem 7 is that verbal feedback istotally confounded with number of exposures to the assessmentitems. Recall that participants in the verbal feedback conditionwere reshown 18 test items, whereas participants in thenonfeedback condition were not reshown the questions. Couldthe effect observed be attributed to the fact that some

participants were simply exposed more often to the questionsthan others were?

This confounding cannot be easily eliminated, per se. Anadditional control group could, however, be added to help ruleout the alternative explanation. A third group could be reshown

the questions but not be given the feedback and explanationabout the correct answer. Comparisons would then be madebetween the three groups to see just how feedback affectsworkplace training.


16/17

8. Okay, now have some fun! In an attempt to show a relationship betweenanxiety-producing conditions and sexual attraction, Dutton and Aron(1974) conducted a field experiment. Male participants were approachedby either an attractive male or female experimenter and asked toparticipate in a psychological study on the effects of exposure to scenicattractions on creative expression. Participants were approached at oneof two locations. One location was a 450-foot-long foot-bridge consisting

of wooden slats supported by wire cables. This bridge was suspended230 feet over a canyon (high bridge). The second location was a solidwood bridge 10 feet over a small, shallow river (low bridge). Participantswere given an item from theThematic Appercept ion Test (TAT) andasked to make up a story about it. After completing the task participantswere told that they could receive more information about the study bycontacting the experimenter directly. At that point the experimenter tore

a small piece of paper from a sheet of paper and wrote his or hertelephone number on it and invited the participant to call. The majormeasure of sexual attraction was the number of participants calling theexperimenter for information. The results showed that fewer participantscalled the male experimenter than called the female and that participantswho were approached on the high bridge were more likely to call thefemale experimenter than those who were approached on the low bridge.


17/17

The confounding factor in this experiment, which as we pointed outbefore was acknowledged by the authors, was a participant selectionbias. It could be argued that participants who choose to cross the moredangerous bridge are more the type who would subsequently call anattractive female than participants who choose to cross the safer bridge.Participants who are willing to take the chance of crossing the highbridge may, for example, be bolder in general, more arousal seeking,

and less inhibited.

Participant selection biases like the one in this problem are nor easilyeliminated. You could, however, conduct follow-up studies to test thelimits of potential alternative explanations. For example, you could haveinterviewed participants 10 or 15 minutes after they crossed each bridge.If you found no differences in participants responses, you could bereasonably sure that the original data were not due to selection biases.You could then attribute the observed differences to the arousal actuallyproduced on the bridge. If, on the other hand, participants still differed,then participant selection might still be a problem. In fact, Dutton andAron (1974) did such follow-up experiments. These experiments supportthe conclusion that heightened anxiety is related to sexual attraction.

Documents

Confounding Exercise - Research Methods (Psychology)