-
Natuarlistic observation
A research method where the researcher will observe the subjects without modifying anything
-
Correlational Approach
The correlational research attempts to determine whether and to what degree, a relationship exists between two or more quantifiable (numerical) variables.
-
Independent Variable
The variable being directly manipulated by the researcher
-
Dependent Variable
The variable that is being studied by the researcher, may or may not be influenced by the IV
-
Between Subjects Design
An experimental design in which each subjects is randomly assigned to only one of the treatment conditions
-
Within Subjects Design
An experimental design where all subjects receive all treatment conditions. Also called a repeated measures design
-
Type 1 Error
This happens when you reject the Null Hypothesis even if it is true.
-
Type 2 error
A Type II Error is also known as a False Negative or Beta Error. This happens when you accept the Null Hypothesis when you should in fact reject it.
-
Confound
A neiscance variable that effects the DV and differs reliably with the IV
-
Factorial Design
Experimental designs in which two or more independent variables are used. This permits the analysis of interactions between variables.
-
Interaction
Occurs when the effect of an IV on the DV differ reliably with the levels of a second IV
-
Nuisance Variables
any variable other than the IV that affects the DV
-
Floor and Ceiling Effects
- Floor - The values of the DV are so low they are not affected by the IV
- Ceiling - The values of the DV are so high they are not affected by the IV
-
Main effect
The effect of the change in level of one factor in a factorial experiment measured independently of other variables.
-
Response Acquiescence Effect
The tendency of an interviewee to agree with the questioner.
-
Random & Fixed factors
- Random - An IV whose levels were chosen randomly from a population of possible values
- Fixed - An IV whose levels were chosen non-randomly
-
Hypothesis
A supposition or proposed explanation made on the basis of limited evidence as a starting point for further investigation.
-
Theory
A supposition or a system of ideas intended to explain something
-
-
Modus Tollens
- If P then Q
- Not Q
- Therefore not P
-
Affirming the Consequent
Using Modeus Ponens backwards (If P then Q, Q, There for P) this form of logic is invalid, based upon this example
- If Bill Gates owns Fort Knox, then he is rich
- Bill Gates is rich.
- Therefore, Bill Gates owns Fort Knox.
-
Levels of the independent variable
values (amounts) of the treatment variable
-
Advantage of the experimental approach
- 1. Causation can be implied
- 2. Precise control of variables
- 3. Experiments can be replicated
- 4. Experiment yields quantitative data
-
Properties of scales of measurement
- 1.Identity Property
- Occurs when different entities receive different scales
- 2.Magnitude Property
- Occurs when the ordering of values on the scale reflects the ordering of the trait being measured
- 3.Equal intervals
- Occurs when a difference of 1 on the scale represents the same amount of the trait being measured everywhere on the scale (e.g. weight in pounds)
- 4.Absolute Zero
- Occurs when a score of 0 indicates a complete absence of the trait being measured
-
4 Types of scales
- Nominal - Identity property
- Ordinal - Identity, Magnitude
- Interval - Identity, magnitude, equal intervals
- Ratio - Identity, magintude, equal inverval, absolute zero
-
Random Selection
Random Selection is a process of gathering (in a truly random way) a representative sample for a particular study
-
Random assignment
Every subject selected has an equal chance of experiencing each level of the IV
-
Quasi-experiment
A study that matches an experimental design in every way except it does not have random assignment
-
Frequency Distribution
A breakdown of how all the scores fell into a set of set ranges (grade distribution)
-
Normal distribution
A function that represents the distribution of many random variables as a symmetrical bell-shaped graph.
-
Positively & Negatively skewed distributions
- Positive - curve towards right
- Negative - curve towards left
-
-
Practice effects
Occurs when the subject's score is influenced by performing a task over and over again
-
Sensitization effects
Sensitization effects occur when the subject is aware of the manipulations used in a study and such awareness causes him/her to change his/her behavior
-
Carry-over effects
results from the previous phase carry-over into the next phase
-
Counter-balancing
the systematic variation of the order of presentation of the levels of the independent variable, used to negate practice & carry over effects
-
Function of a latin square
Used to plan IV exposure of within subjects design & counter-act practice and carry-over effects
-
Internal Validity
How well does something measure what it is intended to measure
-
- level - the chance that one commits a type 1 error
- level - the chance of committing a type 2 error
-
Definition of Statistical power
The power of a statistical test is the probability that the test will reject a false null hypothesis
-
Proving the null hypothesis
Can't be done, you can only disprove it
-
Regression to the mean
Upon retesting extreme scores will migrate towards the mean
-
How the logic of experiments is ruined by confounds
Experiments assume that only the IV can have any effect on the DV and that everything is completely random, however if another variable (confound) has a direct and non-random effect on the DV then one of the base assumptions of an experiment becomes flawed and the experiment fails
-
External validity
How generalizable the results of an experiment are in the real world
-
Demand characteristics
Cues in the experiment that make the participants aware of what the experimenter expects from them
-
Rosenthal Effect
When the subject's performance changes due to subtle changes in the experimenter or their behaviors
-
Placebo Effect
Any effect that seems to be a consequence of administering a placebo, which is a fake treatment or inactive substance
-
Way to overcome placebo effect & rosenthal effect
Run a double blind study
-
Hawthorne Effect
The alteration of behavior by the subjects of a study because they are being observed.
-
Novelty Effect
Occurs when the DV is affected by the IV, not because of any inherent qualities of the IV but rather because the IV is something new
-
Participant observation
A type of research design where the researcher will embed themselves in a group to study them
-
MStreat, MSwithin
Estimate the amount of variance across groups
-
Time series design
A design where data is collected at intervals over a certain period of time
-
Correlation co-efficient
A number between −1 and +1 calculated so as to represent the linear dependence of two variables or sets of data.
-
Proportion of variance accounted for
r2
-
Parametric statistical tests
Statistical tests where certain assumptions about the parameters of the full population of the sample are taken.
-
Survey Research
A research method which uses convienence spampling & paper based questionairres. No randomization however easy way to collect large amounts of data
-
Convenience sampling
A sample where participants are selected at the convenience of the researcher
-
Quota Sampling
A convenience sample with an effort to make sure a certain distribution of demographic selections are made
-
Simple random sampling
A completely random subset of people chosen from a larger population
-
Stratified random sampling
Occurs when a population is first divided into subsets based on some criterion then random sampling occurs within those subsets.
-
Multistage sampling
basically stratified sampling where at each level of stratification there is random sampling.
-
Parts of a manuscript
- Title page
- Abstract
- Method Section
- Results
- Discussions
-
In press
An article that has been accepted for publication but has not been published yet
-
Gallery Proofs
A typeset copy of the article that is sent to the author for final approval
-
How are the Nuremberg code of ethics different from the APA code
The APA code of ethics allows for certain forms of deception to occur to be applied to research subjects
-
Factors determining statistical power
- Alpha level
- Sample size
- Effect size
- One tailed vs Two tailed
- Variance
-
Why do an ANOVA instead of multiple t-tests (2)
- 1. Less paperwork
- 2. Multiple t-tests inflate the type 1 error rate
-
Why can't you imply causation from correlations?
- 1.You don't know the direction of the causal relationship
- 2. You don't know the effects of other variables which are not controlled for
-
Advantages of non-parametric studies
- 1. Less restrictive assumptions
- 2. Often easier to calculate than parametric studies
-
Disadvantages of non-parametric studies
- 1. Tend to be overly conservative
- 2. Some research designs do not have a non-parametric test (factorial designs)
-
Factors to consider when evaluating surveys
- Bias in choosing sample
- Lack of full disclosure when reporting results
- Lying by poll respondents
- Non-respondents influencing the sample
- Characteristics of the poll taker
- Type of questions asked (open vs closed questions)
-
Problems with single subject experiment designs
- 1. Likely to have type 1 & 2 errors
- 2. Placebo effects are likely
- 3. Experimenter bias is a problem
- 4. Generalization difficulties
-
Ethical principles to be considered when conducting psychological research
- Subjects should agree to participate in research
- Subjects should not be coerced
- Subjects should be fully informed about the purpose of the experiment before deciding to participate
- Subject data should be confidential
- Subjects should be warned of any harmful effects of the study
|
|