Analysis of variance can be regarded as a model for explaining the variation of the dependent variable (in this case - the employee productivity organization). Of course, performance depends on many organizational variables. In this example, we consider only two . Analysis of variance indicates that worker productivity differences can be explained by the influence of the variable " level of family income " , as well as the effect of the interaction of the specified variable to the variable " a factor in increasing the motivation ."
Results of the study also allowed to specify the theoretical foundations underlying the development of a new program to stimulate productivity. It can be concluded that, with respect to a specific organizational situation Herzberg motivation theory has serious limitations . It turned out that motivational factors are indeed capable of stimulating the productivity of workers , but the effect is small and applies only to workers with high levels of family income. At the same time for low-income workers a powerful motivational stimulus act hygiene factors and motivators so-called no significant effect . Now , given the identified organizational patterns , the psychologist can develop differentiated performance incentive program that is best for the organization.
Possibilities and limitations of organizational and psychological research
The need for thorough and systematic organizational studies is obvious . However, research conducted in the field, ie, in real organizations , unlike laboratory experiments are often faced with constraints that do not allow to use a variety of research plans , particularly plans number 4 - 7. Most frequently during the fieldwork the following problems occur :
1. Some of the parts of the organization , which is planned to conduct research , for whatever reason, refuse to participate in it. Practice shows that the best way out of this situation is to follow the principle of voluntary cooperation. If the unit or group study perceives as a threat or hindrance of its activities , the scientist and the manager should not insist on their participation in the study.
2 . Complexity and virtually impossible to obtain a random sample of employees and groups in real organizations . For example, selective participation in the study, only some unit employees can not only disorganize his work , but also give rise to an atmosphere of suspicion and hostility. Therefore, management never welcomes selective participation of employees .
3 . Difficulties in the use of the control group . By agreeing to the study , managers certainly want their employees were the experimental group and were directly involved in any research activities . "Passive " participated in the study as a control group perceived as unnecessary and futile exercise , interfere with normal operation .
In circumstances where the use of random sampling is not possible, another frequently used quasi -experimental study design :
Activity measured before the experimental group (O1) and after ( O2) to introduce a new performance evaluation system . The control group (not satisfying the requirements of a random sample ) is estimated from observations of O3 and O4 , carried out simultaneously with the observations O1 and O2. Such a control group will be inconsistent , or not equivalent . This plan is probably the best compromise between the most stringent requirements of the scientific method and specific organizational conditions.
4 . Large load that falls on the psychologist during the study , often difficult objective assessment of changes in the studied variables.
5 . Internal political atmosphere in the organization often has an impact on the course of the study , and the interpretation of its results.
Supporters purity laboratory experiments applied research criticized for the lack of control over many organizational variables. In field studies, the psychologist will inevitably have to move away from the traditional schemes and plans , compliance with which the laboratory strictly regulated . The extent of such waste should be carefully weighed when choosing a researcher of specific methods and experimental schemes.