• Share
  • Email
  • Embed
  • Like
  • Save
  • Private Content
Optimal Internet: Criteria Differences between Effect Sizes, Mediators, and Moderators
 

Optimal Internet: Criteria Differences between Effect Sizes, Mediators, and Moderators

on

  • 365 views

Optimal Internet: Criteria Differences between Effect Sizes, Mediators, and Moderators

Optimal Internet: Criteria Differences between Effect Sizes, Mediators, and Moderators

Statistics

Views

Total Views
365
Views on SlideShare
365
Embed Views
0

Actions

Likes
0
Downloads
1
Comments
0

0 Embeds 0

No embeds

Accessibility

Categories

Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

    Optimal Internet: Criteria Differences between Effect Sizes, Mediators, and Moderators Optimal Internet: Criteria Differences between Effect Sizes, Mediators, and Moderators Document Transcript

    • Optimal Interest 1Optimal Interest: Criteria Differences between Effect Sizes, Mediators, and Moderators by Edgardo Donovan RES 601 – Dr. Roger Rensvold Module 4 – Case Analysis Monday, September 1, 2008
    • Optimal Interest 2 Optimal Interest: Criteria Differences between Effect Sizes, Mediators, and Moderators It has been observed with some truth that "one persons error variance is another persons social behavior"; all models leave certain things unaccounted for, and some leave almost everything unaccounted for. In general, models that leave less variance unaccounted for are preferred to those that leave more, and the tipping point at which a model becomes interesting is the point at which it can account for more variance than can chance alone. ANONYMOUS TUI University, 2008 ffect sizes should be large or small enough to present a pattern of variable variances that enable researchers to work with consistent positively or negatively correlated patterns necessary to prove or disprove an intended thesis. The criteria for mediators andmoderators are different as they are elements that add complexity to a research construct.Moderators have an external influence whereas mediators have an intermediary effectbetween dependent and independent variables. They may either neutralize or enhance therelationship effects between variables. A moderator is a qualitative or quantitative variable that affects the strength of therelationship between independent and dependent variables. Specifically within acorrelational analysis framework, a moderator is a third variable that affects the zero-ordercorrelation between two other variables (Baron). Moderators add complexity to researchconstructs and are often used in research involving leadership (Dionne) andorganizational dynamics but have wider applications as well. For example, one may want tostudy the effect of forced unpaid overtime hours (independent variable) on employee
    • Optimal Interest 3retention (dependent variable) while also taking into consideration the overall familyfriendly corporate policy (moderator) that includes flex-time, tuition assistance, and agenerous 4 week a year vacation package. A goal of such a study could be of determiningthe neutralizing effect of the overall corporate family friendly policy on the negativecorrelation between unpaid overtime work hours and employee retention. Mediators also have an effect on dependent and independent variables in a differentmanner than moderators despite it being uncommon for some people to use both termsinterchangeably. Woodworms model, which recognizes that an active organism intervenesbetween stimulus and response, is perhaps the most generic formulation of a mediationhypothesis (Baron). Simply stated, a mediator affects variables internally akin to agatekeeper. For example, one could attempt to study rude CEO personality trait(independent variable ) effect on employee morale (dependent variable) while taking inconsideration the mediating effect of a kind CEO secretary. A goal of such a study would beto examine the neutralizing effect the secretary has on employee dissatisfaction by
    • Optimal Interest 4containing the negative effect the CEO may have on morale if left unimpeded tocommunicate directly with employees and vice-versa.. In statistics, effect size is a measure of the strength of the relationship between twovariables. As demonstrated in the examples above, effect sizes can be moderated andmediated by other factors which add additional layers of complexity to a research project.The ideal effect size is different for each project. Ideally, the data variance should be in theaccepted range required by the theory it is trying to validate or invalidate. When there are data anomalies, meaning pieces of data that are way outside of thedesired range, this can sometimes be mitigated by increasing the data pool in an attempt tolessen the anomalous impact. Alternatively, qualifiers can be added to the researchconstruct to exclude anomalous data if a proper theoretical reasoning is used. For example,when conducting a study on athletic strength performance in 8 year old boys there may betwo boys out of school of 300 that stand out way beyond the accepted strength range.However, if it is found that one of the boys stayed back 3 years and that the other haddown-syndrome these two could be excluded if the latter criteria were implemented as adisqualifier.
    • Optimal Interest 5 One of the challenges of traditional research, which emphasizes formal hypothesesand significance testing of null hypotheses, is that extreme data variances in the majority ofcases are not desired and can take away from the overall research model applicability. Theexistence of alternative research strategies revolving around large effect sizes and chanceoutcomes coupled with the inability of null-hypothesis testing to guarantee successfulresearch does not justify entirely discrediting null-hypothesis testing thereby throwing outthe baby with the bathwater (Levin). Large effect sizes treat the research of localized anduniversal theories unevenly giving more importance to the latter. Chance outcomes are justthat and are difficult to predict in terms of probability. Operationalizing concepts intoqualitative variables, extending that process into quantitative data-gathering, andconducting null-hypothesis analysis conveys a sense of order to what may otherwise seemas abstract ideas or theories. Effect sizes should be large or small enough to present a pattern of variablevariances that enable researchers to work with consistent positively or negativelycorrelated patterns necessary to prove or disprove an intended thesis. The criteria formediators and moderators are different as they are elements that add complexity to aresearch construct. Moderators have an external influence whereas mediators have anintermediary effect between dependent and independent variables. They may eitherneutralize or enhance the relationship effects between variables.
    • Optimal Interest 6 BibliographyAnonymous. (2008). Effect size. http://en.wikipedia.org/wiki/Effect_sizeBaron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in socialpsychological research: conceptual, strategic, and statistical considerations.. Journal ofPersonality and Social Psychology, 51, 1173–1182.Dionne, Shelley, Yammarino, Francis, Atwater, Leanne James, Lawrence. (2002).Neutralizing substitutes for leadership theory: leadership effects and common-source bias.Journal of Applied Psychology Vol. 87, No. 3, 454–464.Howell, John, Dorfman, Peter, Kerr, Steven. (1986). Moderator variables in leadershipresearch. The Academy of Management Review, Vol. 11, No. 1, pp. 88-102.Frazier, Patricia, Tix, Andrew, Barron, Kenneth. (2004). Testing moderator and mediatoreffects in counseling psychology research. Journal of Counseling Psychology Vol. 51, No. 1,115–134.