Quiz 2 (+ standard deviation notes )

__ are the only method that can be used to prove causation between 2 variables

Experiments

Negatives to experiments

• Results might not be able to be generalized to the real world

• other variables might play a role in the results ( confounding variables )

Causal relationship

Causation/ Cause and effect

One of the main reasons experiments are used is to prove

causation

Experiment set up

• Experimental group receives IV

• control receives no change or placebo

• Group members need to be randomly assigned ( Equal chance to be placed in each group p)

• Double blind study ( neither participants or research know who is receiving the IV

Confounding Variables

Any other variables that might affect the experiment

Researchers try to control these

Ways to reduce bias in a experiment ( list and explain )

• Random Assignment

• SIngle Blind: reduces participant bias by not telling them which group they are in

• Double blind: reduced bias in both participant and experimenter

• Peer review : Allows independent peers to help identify potential sources of experimental bias

Operational Definitions

Detailed definition of your variables to clearly indicate how to measure or quantitfy a concept.

Ex: “Happy” is defined as 5 smiles per 30 minutes.

Naturalistic Observation (explanation + pros and cons )

• Descriptive method of serving in participants natural environment

• Pros : no side effects or responsibility or managing controlled environment because shows in natural environment

• Negative: cannot explain behavior ( because cannot interact with a participant ) AND subjects may act differently if they know they are watched
  

Case Study( explanation + pros and cons )

• Descriptive research where detailed information is gathered over time on a specific individual or small group

• Pros : In depth information AND interaction is allowed

• Negative: Cannot assume a finding is true for all people BECAUSE of the small experimental group ( GENERALIZATION ) AND often time consuming

Meta-Analysis ( Explanation + Pros and Cons )

• Statistical procedure for analyzing the results from 2 + separate studies to reach an overall conclusion

• Pros: Increase precision in findings AND can settle controveries/conflicting claims

• Negatives: not using new information and may be missing potentially useful information from original research if not reported ( something may not be relevant to original study but is relevant to meta analysis )

Correlation Study ( Explanation + Pros and Cons )

• Looking at 2 variables to see if there is a statistical connection/relationship

• You can gather data from any of the other types of research

• Positive: Correlation proves some sort of relationship ( or lack thereof ) statistically

• Negatives: CORRELATION DOES NOT PROVE CAUSATION ( OTHER FACTORS COULD BE AT PLAY )

Correlation Coefficient ( definition, important things to note )

• Shows the type of relationship and strength between 2 variables

• -1.0 = strongest NEGATIVE relationship ( Inverse relation )

• 0 = no relation

• +1.0 = strongest POSITIVE ( direct )

• When looking at strength of correlation coefficient, ignore the sign at look at if the number is closest to 1

• Graph data on a scatter plot

• 1 variable on x axis and 2 variables on y axis

What is the strongest strength based on correlation coefficient ?
A ) + 0.6 B) -0.85 C)+ 0.2 D) -0.9

D, because it is the closest to 1 when you only look at magnitude

Ex Post Facto

( after the fact ) , looking back at something after the fact

Longitudinal Study

• Looks at the same group of people at different points of their lives ( takes many years )

Cross Sectional

• Looks at people at a variety of ages all at the same time ( only happens at one period in time )

Survey

• collect attitudes, optionion, or behaviors by asking questions to a representative sample of a certain population

How to pick a simple for a survey?

• Identify relevant populations

• Random Selections ( every member of population has an equal chance of being chosen )

Survey concerns ( List and Explanations )

• Self-reporting Bias : People give inaccurate answers often due to social desirability

• Sampling Bias : Flaw in sampling skews data

• Framing: the wording used for questions influence the interpretation and response to them

All research must pass a review conducted by…. to determine..

Institutional Review Board ( IRB ) to determine whether or not the study is ethical and poses a risk to those involved before it is conducted

In research specific criteria must be followed as set by the

American Psychology Association

Goals of ethical guidelines ( 3 )

• Ensure benefits outweigh risks

• Maintain concern and respect for well being and autonomy of individuals

• Ensure fair procedures

Ethics codes ( 5 )

1. Obtain voluntary informed consent ( includes option to withdraw, for minors they give ASSENT )

2. Protect participants from unnecessary harm of discomfot

3. Protect confidentiality of all participants

4. Debrief individuals afterward ( explain results and indicate if there was any deception )

5. Receive approval from Institutional Review Board

Ethics for Animal Experiment ( 3 )

• Provide humane conditions

  • minimize pain and discomfort

  • use anesthesia whenever possible

  • veterinarians must be consulted whenever using live animals

• Can prove benefits of society outweigh costs ( pain on animal )

• Have planned end date for study

Qualitative Research ( relies on… + example )

-Relies on in depth , narrative data ( not numbers)

Ex: Structured interviews

Quantitative Research ( Relies on… + Example )

• Relies on quantifiable , numerical data

• example: Likert Scale

Likert Scale

• Linear or continuum scale to measure attitudes

• Improve operational definitions

• Strongly Disagree = (1) , Disagree ( 2 ) , Undecided ( 3) , Agree ( 4 ) , Strongly Agree ( 5 )

Measures of central tendency , ideal

Mean , median, and mode.
Ideally these are all the same number to get the standard bell curve standard deviation

When scores are closer in range, the standard deviation curve …

will appear skinnier / taller and have a smaller standard deviation

When scores are more spread out the curve..

will be shorter/fatter and have a larger standard deviation

Normal vs skewed distribution ( explanation , not a list of skewed distribution )

• Normal = mean , median, and mode are all this ame

• Skewed = extreme scores or when mean median and mode are not the same value

When it comes to skewed Distribution , the __ is a better measure of central tendency than the _

Median, mean

Types of skewed Distributions

• Positive

• Group has one high score and it contains more low scores

• Negative

• One low outlier and there is more high scores

Bimodel Distribution

• 2 different values appear most frequently in the data set

• AKA 2 diff modes

• creates a 2 hump curve

•