Name two other variables that need to be controlled in this study and describe how you would control them. Time of day must be controlled. All subjects should be tested in the same day-part, and once in the morning and once in the evening, because cognitive abilities cycle and change in strength during a 24-hour period. Noise in the experiment area must also be controlled and kept to a minimum. d. Describe the procedure. What will happen in the experiment from the time the subject enters the testing room until the time the subject leaves.

There will be no talking except “Hello”, “Goodbye”, and “Thank You” in the experiment room, between subjects and experimenters. The first subject enters the room and sits down at the table. The experimenter (#1) has already shuffled the cards (including new decks) with the card shuffler before the subject entered. The experimenter places each card face down one at a time in front of the subject for 8 seconds each. The subject is given 8 seconds to respond with the name of the unseen symbol.

Longer times and “do-overs” will not be permitted. Each card that is responded to or not is slid over to the right of the experiment and the next used card will be placed on top of it, still face down, forming a stack of 25 at the end of the test. Experimenter #1 will not touch that stack again. Each subject will go through 25 cards one time per test session. The experimenter records each response on a chart for each subject for the 25 cards. Each inability to respond will be counted as an incorrect answer.

The completed face-down card stack will remain untouched until the subject leaves the room. A second experimenter will come out of another room and take the deck into the second room and turn over the entire stack and then check whether responses were correct or incorrect on the subject’s Record Chart received from the first experimenter, and then file it while the first experimenter shuffles another deck of cards or picks up one already shuffled. The first deck of cards may be used for the 3rd subject, the 2nd for the 4th subject, and so on.

When a card is defaced in any way, the entire deck will be discarded after the associated subject’s answers have been checked, or immediately if defacement occurs between subjects or at other times. If the experimenter drops a card face up in front of the subject, that card is discarded and one taken from an extra, new, shuffled deck kept on hand separately in the second room. e. What is the independent variable or variables in the study? What is the dependent variable? The independent variable is the day-part of testing. The dependent variable is the % correct score. 3. Results a.

Describe how you would perform a statistical test on the experiment you designed for Dr. Venkman. What groups would you compare? How would you determine the difference between groups? A t-test that compares mean correct % score will be done between subject and control groups for 1) each day-part and 2) the totals for each group. Group differences will be determined with associated significance procedures in order to arrive at p-scores, and data may be further examined through an ANOVA, although sample size is small and significance is likely not to be garnered through the more sophisticated program.

b. How would you determine the variation within groups? Variance and standard deviation will be calculated for each group for each day-parts and total. c. Define statistical significance. What would lead you to conclude that the difference between groups is statistically significant? Statistical significance is the probability that the difference between groups found is substantially related to factors other than chance. If a p-score is < 0. 05, then the difference between groups is more probably statistically significant. A p-score < 0.

1 is more highly significant, and so on. If someone says that there exists one chance in a thousand something occurs by chance, they are proclaiming a 0. 1% level of statistical significance. 4. Conclusions a. Would Dr. Venkman be correct in concluding that both psychics and non-psychics identify Zener cards better than would be expected by chance (4 subjects form each group)? Support your position. No. Testing only four subjects in each group is not using a sufficient number in order to determine this. The variance among four subjects is too high in possibility.

For example, one subject could score 100% and the other three, much lower. b. How is it possible that the participants in the control group actually have better accuracy than participants in the psychic group? This is solely a function of the low number of subjects considered, combined with chance. c. What are the most appropriate conclusions that can be drawn from the results depicted in the figure? What does the result imply about the reality of clairvoyance? Examining only four subjects in each group is insufficient to draw any conclusions or to state anything new about the reality of clairvoyance.

The only fact in evidence is that anyone can be correct 5% of the time. d. What recommendations would you make for improving the study? An accurate and appropriate sample size must first be determined, based on results and variances from previous related research, then entering the necessary statistics into a standard sample-size equation. Within the larger sample/group sizes, between-group matching can occur for age, gender, and race. Additional day-parts may be added, including late-night testing.

References

Myers, D. G. (2007). Psychology (8th Edition). New York: Worth.