My first hypothesis is observed fairness of punishment outcomes for a given incident of misconduct, a subordinate is expected to receive punishment similar to that which others have received will be positively related to no repeat of such misconduct. In other words, I will like to test the equity.
Equity is something that is just, impartial and fair. Procedural justice is related to the effects of punishment systems on cooperation. (Science Letter 2008) Punishment is an outcome of a decision-making process (e.g. a court trial), and it stands to reason that the perceived fairness of this process can vary systematically. Therefore I will set up procedures of punishment in the experiments with or without logic and then provide to the data subjects. The logic of punishment enforcement is set up to create a consistent or an inconsistent situation. People evaluate procedures as fairer than if procedures are inconsistent (Trevino 1992). Therefore, the effect of fairness can be measured by setting up consistent and inconsistent situations. That is, I will set equity as a nominal variable as observed consistent punishment / observed inconsistent punishment.
Other than the perception of equity, I would also like to test the contingency. An employee does not repeat misconduct may also be related he/she agrees such misconduct is inappropriate behavior which should not be repeated. Therefore, my second hypothesis is observed inappropriate behavior being punished will be positively related to no repeat of such inappropriate behavior.
Contingency is a something incidental to something else. Contingent punishment behavior can be described as people are perceived to getting disapproval, displeasure and administer reprimands for poor performance. (Jiing-Lih 1987) And generally people respect for the dignity of humans (Welton 1993), so I will relate ethical and unethical behavior to contingent punishment has been enforced or not. So as to measure the effect of contingency, I will let the data subject perceive that no punishment will be enforced to unethical behavior which is a condition of noncontingent punishment. That is I will also set contingency as a nominal variable as observed contingent punishment / observed noncontingent punishment.
Repetition of others misconduct involve the duration of measurement. I will measure the carefulness of the data subject of their preventing not to repeat the same mistakes made by others before and after their notification of different setting of enforcement (if any) of punishment. The repetition will be measured by the number of mistakes they have made and the change of their attitudes as reported by them. That is, I will set repetition as an interval variable by measuring the number of mistakes which permits calculations. At the same time I will also set the awareness of repetition as ordinal variable by ranking made by data subject's level of carefulness (less careful / as careful as the last test / more careful) to avoid repeating other's mistakes.
To address the research question guiding this study, I will set up laboratory experiments which are excellent for controlling variables. So as to improve the realism and avoid the effect of perception of punishment to be carried out in the organization on the subjects, I will randomly divide each company's employees into five groups (A, B, C, D, E) and let them attend typing speed test in different rooms with the same furniture settings while they are not allowed to communicate to other groups until the end of the experiments because the information they will be told during the experiment does not really happen. That means the data subject will perceive punishment (if any) is to be enforced on their known colleagues. All of them will be requested to fill in a questionnaire regarding their memory regarding the latest disciplinary action that the company has taken and their attitudes and response towards the company's action. The facts will be compared and analyzed as background information so as to see if there is any history effect.
Is consistent procedural punishment enforcement positively related to no repeat of such misconduct? After the first typing test, group A will be provided a record of their colleagues (group B and C)' mistakes made in the typing test plus the logic of disciplinary actions was taken while group B will be provided with a record of their colleagues (group A and C)' mistakes made in the typing test plus a record of different disciplinary actions are randomly assigned to the list of colleagues who made typing mistakes. And, group C will be provided with a name list of their colleagues (group A and B) so that they cannot aware that the typing test will be collected with punishment (if any) on their colleagues while group C is the control group. And then, all of them will be requested to attend the second typing speed test with the same level of difficulties and fill in a questionnaire about their feeling (less careful / as careful as the last test / more careful) and prediction of the typing test result (less mistakes / the same / more mistakes) on the same day. Their feeling is collected for further calculation and test while their prediction of the typing test result is for counter checking of the consistency of their feeling. It is because if the data subject completes the test more carefully should expect less mistakes he / she has made. I expect to learn their reasons for any changes of the attitude toward complete the first and second typing tests so as to find out the influence of consistencies / inconsistencies of punishment reinforcement.From the above experiment, I will collect the following data by the following procedures:
- Select group randomly:
- Group A and B are experimental groups, group C is control group.
- Give 1st typing test (pretest) to all groups:
- No. of mistakes made by group A is OA(1), that by group B is OB(1), that by C is O(3)
- Provide the punishment enforcement information (X)
- Give 2nd typing test (posttest) to all groups:
- No. of mistakes made by group A is OA(2) that by group B is OB(2), that by C is O(4)
- Calculate experimental group score for group A (observed consistent punishment)
- SA(E) = OA(2) OA(1)
- This is improvement over time (post course) of the experimental group
- Calculate experimental group score for group B (observed inconsistent punishment)
- SB(E) = OB(2) OB(1)
- This is improvement over time (post course) of the experimental group
- Calculate control group score (observed no punishment)
- S(C) = O(4) O(3)
- This is improvement over time (without observed punishment) of the control group
- If SA(E) is significantly larger than S(C), then observed consistent punishment is effective.
- If SB(E) is significantly larger than S(C), then observed inconsistent punishment is effective.
- The frequency of "more careful collected from group A is OA(5)
- The frequency of "more careful collected from group C is C(5)
- If OA(5) is significantly larger C(5), then reject the null hypothesis and conclude that observed consistent punishment is positively raise the awareness of no repeat of other's mistakes.
From the above experiment, I will also collect the following data by the following procedures:
For the comparison of group B and C, I will take "less careful into consideration.
Is contingent punishment enforcement positively related to no repeat of such misconduct? After the second typing test, group D and E will be told to enter the third typing test which aims no mistakes instead of high speed and they will be told additionally that their colleagues (group D or E and C) have been cheating such as using the spell check function without approval before handing in paper in the second typing test. Group D will be remarked that no punishment was made for those behaviors, while group E will be remarked those dishonest people were disqualified. And, group C will remain the control group and will be remarked nothing. After the typing accuracy test, the logs of the computer used by group C, D and E will be checked and identified who has committed cheating in the third typing test but not in the second typing test.
From the above experiment, I will collect the following data by the following procedures:
- The number of data subjects belongs to Group D committed cheating is OD(6)
- The number of data subjects belongs to Group C committed cheating is C(6)
- If D(6) is significantly larger C(6), then reject the null hypothesis and conclude that observed noncontingent punishment enforcement negatively related to no repeat of such misconduct.
For the comparison of group E and C, I will be able to test contingent punishment enforcement positively related to no repeat of such misconduct if E(6) is significantly smaller than C(6).
All the above experiments will be repeated in different companies to test the generalizability of the hypothesis.
- Anonymous., Aug 5, 2008. Life Sciences; New life sciences study results from Free University described, Science Letter. Atlanta: pg. 2748.
- Jiing-Lih Farh, Philip M. Podsakoff, Bor-Shiuan Cheng, Fall 1987, Cuture-Free Leadership Effectiveness Versus Mederators of Leadership Behavior: An extension and Test of Kerr and Jermier's "Substitutes for Leadership Model in Taiwan, Journal of International Business Studies. Washington: Vol. 18, Iss. 3.
- Trevino, Linda Klebe, Ball, Gail A., Dec 1992, The Social Implications of Punishing Unethical Behavior: Observers' Cognitive and Affective Reactions, Journal of Management. Bloomington: Vol. 18, Iss. 4.
- Welton, Ralph E., Davis, James R., Winter 1993, Understanding Ethics Development and Employee Behavior, Boston: Vol. 8, Iss. 3.