Research in OB Flashcards
What are the advantages of evidence based management?
Removes bias; you will be more likely to get the results that you actually want. Businesses are able to better predict what kind of effect their decisions will have
Why is evidence based management more likely to get desired results?
Research indicates the best practices; what tends to work, what has worked for other companies, etc. and you have proof that it is likely to work
What is research?
Finding out about the world through objective and systematic observation/information gathering
How is OB separate from personal opinions?
OB researchers follow strict methods to conduct quality research; this separates OB from opinion, intuition, and common sense, which are more likely to be biased and inaccurate
Explain how our intuitions can often be wrong
Because of own experiences, we can become overconfident in our ability to understand other people and how expect them to react to different situations in the workplace. Often these intuitions are wrong
Define empirical information:
Known through observation; must be verifiable by other people using same methods
Define falsifiable information
There must be a way to show empirically that a claim is NOT correct
What is the process of conducting OB research?
- Decide of your research/topic
- Review literature
- Articulate your theory and form your hypothesis
- Design study
- Collect data
- Analyze data and report findings
- Apply the findings (eg. define different actions you can take to improve a situation)
Define hypothesis
A formal statement of the expected relationship between two or more variables
Define theory
A set of statements about the inter-relationships between concepts that allow us to predict and explain various processes and outcomes. Must be testable
What is a moderator?
A moderator is a variable that changes the relationship between two other variables (eg. age: how a rude managerial style affects employees’ satisfaction/productivity depends on the employees’ age)
What is a mediator?
A mediator is a variable that explains the relationship between two other variables
What are the three types of research?
Observation, correlation, experimentation
What is participant observation?
When the researcher becomes in insider of the group; they become a functional member of the group that is being studied
What is one caution about participant observation?
By being part of the group, there is a possibility that you can change their natural behaviour (either positively or negatively). This can invalidate your study
What is direct observation?
When the research only watches from the outside of the group; they have no involvement in or with the group being observed
What are the advantages of observational design?
Depth, breadth, richness, spontaneity, realism. You can get a lot more information about people by watching them in their natural environment rather than asking them questions directly
What are the disadvantages of observational design?
Time and effort-intensive, lack of control, small number of cases, limited generalizability to other settings. More bias can be created if researchers don’t notice something/ignore something that might be important/pay more attention to things that aren’t as important
What is meant by “continuum of designs?”
The three different research techniques can be used in combination or in series to complement each other
What are the two types of observational research?
Participant, direct
What are the advantages about correlational design?
It is the easiest technique to conduct; you can gather much larger samples from the population. You can take larger samples because the methods are more efficient
What are the three methods of collecting data using correlational techniques?
- Survey
- Interview
- Existing data
Why is it important to have large samples?
It helps to bump out any outliers. You increase the probability that the people you are studying are an accurate representation of the actual population. It gives us more confidence in our findings
What information do we get from correlational design?
If there is a relationship between two variables and if it is negative or positive
What is the equation for a positive relation?
r>0
What is the equation for a negative relation?
r<0
What is the equation for an equal relation?
1:1
What is the equation for a non-correlated relation?
r=0
If two variables are correlated, does that mean that one causes the other?
No, correlation does not mean causation. There might be other factor’s that can influence the outcome
What are some examples of correlation using variables?
x is related to y, x and y are associated, x and y change together
What are some examples of causation using variables?
x leads to y, y results from x, x predicts y: x changes first and y changes along with x
What are the key differences (things to remember) between correlation and causation?
Knowing the correlation does not mean that you know the causal relationship; knowing the correlation does not mean that you know why an association exists
Define an experiment
When the independent variable is manipulated or changed. The dependent variable is not changed - we think that the dependent variable will depend on the independent variable
What must you have in order to ensure the success of an experiment?
You must have additional controls (thereby increasing your internal validity), because some other factor could have caused the improvements we observe
Can we guarantee that the experiment results are a direct cause of the experiment?
No - but we can strongly suspect and have compelling evidence
What can you do to make your experiment as accurate as possible and isolate the variable?
- Use a control group
- Control all other factors at the same level (personality, age, experience, etc.)
- Random assignment - increase the chances that the groups are equivalent in other factors
What are two key factors of research?
Reliability, validity
Describe reliability
In order for your research to be reliable, you have to be able to replicate it
Describe validity
In order for your research to be valid, it has to measure what you want it to measure and not some third variable
You can have any possible combination of reliability and validity EXCEPT:
It is not possible to have something that is valid but not reliable
Define internal validity
The extent to which a researcher can be confident that there are not alternative explanations for observed relationships: is it saying what you think it’s saying? is it measuring what you think it’s measuring?
Define external validity
The extent to which the results of a study can be generalized to other samples and settings: will it apply to the more general population, apart from the people you are studying? are your samples too small/too specific to reflect the general population?
Define the Hawthorne effect
When research subjects change their behaviour because they know they are being watched
What is one thing you have to be cautious of in research related to the Hawthorn effect?
The relationships you find in your study might not be true relationships because people’s responses changed as they knew they’re being watched
How can you lower the threat to internal validity?
Random sampling
Define random sampling
Research participants have been randomly chosen from the population of interest
