Appendix

Hypothesis - formal statement of expected relationship between two variables

Variables - measures that can have 2 or more values

  • Independent variable - a predictor or cause of variation in a dependent varibale.

  • Dependent variable - a variable that will vary with changes in the independent variable.

  • Concepts - The idea of a relationship between variables.

  • Operationalizations - The act of taking concepts and making them into something concrete and measureable.

  • Moderating variable - variable that affects the strength of the relationship between an independent and a dependent variable

    • Ex. country of origin, gender.

  • Mediating variable - intervenes or explains the relationship between an independent and dependent variable.

    • Ex. presence of a certain variable

  • Reliability - an index of consistency of a research subject’s responses.

  • Validity - an index of the extent to which a measure reflects what it is supposed to measure.

    • Convergent validity - where there is a strong relationship between different measures of the same variable.

    • Discriminant validity - a weak relationship between measures of different variables.

Observational research - research examining the natural activities of people in an oragnized setting by listening to what they say and watching what they do.

  • Participant observation - the researcher becomes a functioning member of the organizational unit.

  • Direct observation - the researcher observes organizational behaviour without participation in the activity being observed.

    • Research process:

      • Research question > form hypothesis > design a study > collect data > analyze data and report findings > evidence-based management

Correlational research - measuring variables precisely and examining relationships among these variables without changing the research setting.

  • Correlational design - passive observational data, observing the naturally occuring relationship with two variables.

    • Surveys, interviews, existing datas.

    • Scatter plot

    • Range from -1 to 1

      • Correlation of 1.0 is a concise line, .5 is moderate amount of scatter, .2 more scatter.

    • Results look like: x is related to y, x and y are associated/x and y change together.

      • In contrast, causal relationships usually are x leads to y, x predicts y.

      • Correlations do not equal causation.

    • Spurious correlation - if you measure enough things, they can covary by chance.

      • Ex. chicken consumption and total US crude oil imports.

      • There often exists a third variable that increases both of the variables.

  • Cross-sectional design - when the independent and dependent variables are measured at the same time.

  • Longitudinal design - the independent varibale is measured at one point and the dependent at a later point.

Expirimental research - manipulating nature; a variable is manipulated or changed under controlled conditions.

  • Researcher is taking control of the cause; manipulates the cause (IV) the consquence (DV) of the manipulation for some other variable is measured.

  • 2 essential features:

    • Manipulation of the independent variable

    • Random assignment to condition

      • Has a purpose of overcoming the third variable problem by distributing all other variables equally among the experimental groups.

      • Makes the groups equivalent to rule out the possibility that other factors can cause both of the factors to correlate.

    • Random assignment - of participants to experimental and control conditions, improves internal validity.

      • Internal validity - the extent a researcher can be confident that changes in a dependent variable are due to the independent variable.

        • Higher in experimental research.

        • Threats to internal validity - factors that are alternative explanations for the results of an expiriment.

      • External validity - the extent to which results generalize beyond current sample, setting, etc.

        • Higher in correlational research.

  • Control group - a group of research participants who have not been exposed to the experimental treatment.

  • Quasi-expirimental design - when partifipants cannot be randomly assigned, and variables are measured that they might differ on and then compared on those variables to control for any differences.

Three issues and concerns in organizational behaviour research

  • Sampling

    • Generalizing results beyond their study

    • External validity - the extent to which the results of a study generalize to other samples and settings.

    • Random sampling - research participants are reandomly chosen from the population of interest.

  • Hawthorne effects

    • The favourable response of participants in an organizational expiriment to a factor other than the independent variable that is being manipulated.

  • Ethics

    • Researchers must avoid unecessary deception, inform participants of the general purpose of the research, and protect anonymity of research participants.