Types of Validity & threats
Experiments are internally valid when the changes you see in a study, were really caused by the thing you tested and not by some other factor.
Internal Validity
Numbers that represent the variable's qualities or categories
Ex: Female and Male
Nominal
All individuals have an equal , fair chance of being selected.
Random sampling
The variable you manipulate
Independent variable
A research setup used to test if there is a significant difference between three or more groups based on one factor
ex: Measuring three types of therapist (gender, kind of trainning) to see if there is any difference in the clients self-esteem or are we looking at the interaction between gender and tranning
On way Anova
Because we are looking for the effects of two or more factors (their gender, trainning etc.)
Relates to how these findings can be applied to the real world
External Validity
The numbers represent the differences of the variables throughout the scale, you can ORDER the data from top to bottom
Ordinal
Selecting in such a way that major subgroups (ethnicity, gender, age etc.) in the population will be sample.
Stratified sampling
Ex: men and women are broken into smaller groups then are pickly randomly to ensure both are represented
What does it mean when you have a Type I error?
Referring to the rejection of the null hypothesis (which states that there is no difference) when it is correct. In short you make an error
When your research involves more then one dependent variable.
ex: measuring self esteem and a person belief
Multivariate (manova)
The response of the subjects may be influenced by the researcher ex: treating subjects differently or reinforcing different behaviors
Experimenter bias
The intervals (distance) between the numbers on a scale contain the same amount of variables throughout the scale
Ex: the distance between 11 and 12 is the same distance as 24 and 25
Interval
Selecting groups of individuals such as classrooms, city blocks etc.
Cluster sampling
sampling groups instead of indidviduals
Name the IV and DV in the example.
The effect of three kinds of techniques on anxiety.
IV kinds of techniques
DV Anxiety
The effect of any pretest used on the experimental treatment.
ex: FOUR groups take a pretest (A,B,C,D), they then go through treatment and the same FOUR groups take a postest.
This design allows you to determine whether the pretest by itself made a difference, the treatment by itself made a difference or was it a combination of both. OR if nothing actually made a difference. hmmm
Solomon four-group design
Difference in results may be due to instruments that are unreliable or are changed during the study
Instrumentation
The numbers are on a scale which has true zero
Ex: Someone who weighs 200lbs is twice as heavy as someone who is 100lbs
Ratio
Selecting the same proportion of individuals for the sample to accurately reflect the makeup of a larger popultaion.
Proportional stratified sampling
*you want to make sure the sample has the exact same percentages as the real world. Instead of randomly picking people and hoping for a good mix you intentionally pick a smaller group that is a "mini-me" version of the population.
What is a Type II error?
Failure to reject the null hypothesis when there is, in fact a difference.
ex: curious if whether more boys than girls wear jeans to highschool
Chi square
Differences among participants at the start of the study (the way they are chosen)
Ex: comparing two groups, one with a new teaching vs an old teaching method. Groups are not equal
Selection of subjects
Gender and Religion are both examples of what type of level of measure and why?
Nominal- which is data that names or labels categories without any order or ranking
Specific number of participants, items or observations included in a research study
The variable you are measuring or trying to change
Dependent variable
When neither the researcher or the subject knows who is getting the active substance or placebo.
Double-blind techniques