Research and Program Eval Flashcards
Most valuable type of research is
the experiment, used to discover cause-and-effect.
Internal validity refers to
whether the dependent variable (DV) were influenced by the experimental IVs (independent variables) or other factors impacted it.
External validity refers to
whether the experimental research results can be generalized to a larger pop. (people, settings, or conditions).
Experiments emphasize parsimony
this means interpreting results in the simplest way.
Occam’s Razor suggests that experimenters
interpret results in the simplest manner.
a counselor is running an experiment to test a new form of counseling. Little do they know one client in the study is secretly seeing a Gestalt therapist. the experiment
confounder/ flawed
nondirective is to person-centered as…
parsimony is to Occam’s Razor
what publisher produces the most counseling research journals?
APA’s Journal of Counseling Psychology
An experiment is confounded when…
an undesirable variable (aka the contaminating variable) are not kept out of the experiment
Basic research is done to…
advance understanding of a theory
Applied research (aka action research)
conducted to advance knowledge of theories, skills, and techniques in practical application.
IV is the variable
researchers manipulates, control, alters, or wish to experiment with.
“I” control
The DV expresses
outcome or the data
counselor hypothesized that biofeedback (true experiment) training could reduce anxiety and improve average score of written board exams. If this professor decides to conduct a formal experiment the IV will be_____and the DV will be________.
biofeedback; board exam.
experimenters should always abide by the code of ethics. The variable you control in an experiment is??
Independent Variable/ IV
To conduct an experiment regarding a hypothesis you need______and______.
a control group with at least 60 subjects; experimental group.
quasi-experiment
when it is not randomly assigned to two groups
Hypothesis testing is closely related to the work of…
R.A. Fisher
hypothesis
a statement that can be tested
What is a Null Hypothesis?
The IV does not impact the DV
What is an alternate/ affirmative hypothesis?
IV has caused a change
To compare the control group and experimental group the researcher needs
test significance, if the results significant or due to chance.
percentile rank
descriptive stat that tells you where the percentage fell below a certain level
ex: if joe’s score is at 50th percentile then 50% of ppl had raw scores below him.
percentage score
a raw score. ex: percentage of 50 could be low, high, or average. it means half of the test is answered right.
the letter “P” relating to test significance means
probablity.
parameter
a value obtained from a pop of stats drawn from a sample.
in social science the accepted probability level is…
.05 or less.
(most popular are .05 and .01)
P=.05 means
a. there is only a 5% chance difference btw the control group and the experimental group is due to chance factors.
or
b. differences truly exist; the experimenter will obtain the same results 95 times out of 100.
P=____ would rule out chance factor
.001 (the smaller the p level the less significance)
Type I and Type II errors are called ______and______.
alpha; beta.
A Type I error occurs when
you reject null when false. (reject when true since it is possible it happened by chance)
Type II error
is called a beta error, this means you accept null when it is false.
experiment results indicated that biofeedback helped raise written board exams BUT in reality that is not the case.
Type I error (reject null hypothesis).
researcher increases their sample size in the experiment. This will…
Reduce Type I and II errors
researcher changes significance level from .05 to .001…
alpha error decreases; beta error increases.
(lowers from .01 to .001 lowers the risk of Type I errors but increases Type II.)
to test the significance of the differences btw the groups the counselor utilized
the student’s t-test (analysis of variance).
-if the t-value obtained stat is lower the critical-t you accept the null hypothesis
when is an ANOVA (one-way analysis of variance) utilized?
when there is more than 1 level of a single IV.
if a researcher utilized 2 IVs then the stat of choice would be…
Two-way ANOVA or MANOVA
to complete a t-test, consult a tabled value of “t”. To see if significant differences exist in an ANOVA you would…
table for F values
what level of significance would rule out the chance factor?
.001
what is a correlation coefficient?
a stat that shows a degree or magnitude of a relationship btw two variable (abbreviated as “r”)
True or false: A positive correlation (both variables in the same direction) a stronger relationship than a negative (variables go opposite ways) one?
False, a correlation of -.70 shows a stronger relationship than a positive correlation of .60. The minus sign just shows if the variable goes up or down.
Which of the following most like yields a perfect correlation of 1.00?
length in inches and length in cm.
Correlation does not mean causal!!
correlation research is quasi-experimental and does not yield cause-and-effect data
N (number of ppl)=1, intensive experiment design…1st step would be…
take baseline measure
single-blind study
in a study where clients do not know who is a part of the treatment group or control group
double-blind study
study where the clients and researchers do not know who is apart of the treatment group or control group.
correlational coefficient
descriptive stat that indicates the degree of “linear relationship”. (1.0 or -1.0)
In a normal curve, the shape is_____ bc the mean, median, and mode all fall right into the middle of the curve.
a symmetrical bell
68-95-99.7 Rule(empirical rule)
normal distribution of 68% of score fall within +/- standard deviation (SD) of the means; 95% within 2 SD of the mean; 99.7% within 3 SDs of the mean.
