Causal Comparative design
describes a true experiment except for the fact that the groups were not randomly assigned but can be anaylized with a test of significance just like a true experiment.
Ex post facto or after the fact design
can be analized with a test of significance-t-test or anova-just like a true experiment.
Independent variable
I experiment with
Dependent variable
is the outcome
Ethnographic research
collected via interviews, observations and inspection of documents
t-test
simplistic form of the analysis of variance is used to see or asertain if two sample means are significant
When a t-test is computed it will
yield a t-value found in the back of book table.
t-test computation must
exceed the number in the table in back of book to reject null
t-value in back of book is sometimes
called critical t
t-test if more than two groups
then the analysis of variance is utilized (ANOVA)
ANOVA- when used to compare more than two sample yields
and F-statistic
F-statistic--consult back of book for F table
if value of F exceeds the critical value in back of book reject the null
Correlational data (bivariate)
two variables
Corrlational data (multivarial)
three variables or more than two
Baseline measure (before treatment)
Treatment is implemented and behavior is measured again
A s signify baseline, B s
signify intervention implementation and C s are second or alternation form of intervention
Idiographic studies
often single case investigation are called ( sometimes misleading and not necessarily generalizable)
Factorial design
used when there are two or more independent variables
Harmonic mean
compute 2+2+4 and convert to 1/2+1/2+1/4 (add) which =1 and 1/4 or 1.25/3=.4166
Benefits of standard scores- such percentiles, t-scores, z-scores, stanines or standard deviation over raw scores are
allows you to analyze the data in relation to the bell curve.
Independent variable
use horizonal axis (x axis) to plot
Dependent variable
the outcome factor; the variable that may change in response to manipulations of the independent variable -use vertical axis (Y axis)
Inclusive range
use + 1 when substracting between highest and lowest score for range
Variance
standard deviation squared
z-score
same as standard deviation
T-score
use mean of 50 with each standard deviation as 10
Parametric data
rely strictly on interval or ratio data
Non-parametric
tests are designed for nominal or ordinal information
Nonimal scale
has no zero point and does not indicate order (eg. street address, telephone #, political party affilication, gender, brand of thrapy--adding, subtracting, multiplyiing or dividing would be meaningless.
Ordinal
provides relative placement or standing, but does not delinate absolute difference-again adding subtracting, or multiplying would be meaningless.
Interval scale
numbered scale at= distances but has no 0 point-can add or subtract but cannot multiply or divide.
Ratio scale
is an interval scale with a 0 point--adding subtracting, multiplying, or division are possible. Time, height, weight, temperature on Kelvin scale, volume, distance meet requirements.
Hawthorne effect
person not receving counseling, but observed in a study are improving. If subjects know they are part of an experiment and given more attention--their performance improves
Rosenthal effect
experimenter's belief about individual may cause individual to be treated in certain way.--Individual begins to fulfill experimenter's expectations.
Halo effect
a trait not being measured influences researcher's rating on another trait
Horizonal sampling
selects subjects from a single socioeconomical group
Vertical sampling
persons from two or more socioeconomical classes are utilized
Systematic sampling
take every nth person (first person out of every ten)
Operation definition
outlines a procedure so other reseaerchers can attemp to replicate
Inductive
research goes from the specific to a generalization
Deductive
reduces the general to the specific
Parsimony or Ocam's Razor
interpret results in simplest manner
Parmeter
a value obtained from a population (e.g., male'height