Problem Set #2 Terms Flashcards
Alpha
This is the confidence level we use in expressing significance.
We use α=.05
To express significance in our interpretation as p
T/F α=.05 is the historical standard (Winzenz)
True
What is the alternative hypothesis?
The hypothesis we hope to support in our research, predicting that a significant relation or difference exists between the groups we are comparing.
What is known as the opposite of Null Hypothesis?
Alternative Hypothesis
In the analysis decision tree explain what if mean if you are only comparing one group
- a one sample t test
- measures the mean and standard deviation
In the analysis decision tree explain what if mean if you are comparing two groups…
- r or two-tailed t-test or Wilcoxon
- R test measures DV & DV
- Two-tailed t-test & Wilcoxon measures IV & DV
Define Citation
Stat information that appears at the end of any conclusion or interpretation.
t(14) = 2.78, p
Cohen’s convention for describing effect, strength, or variance
.0 - negligence, very little, or none (nearly total overlap (90%))
.2 - weak, small, not much (80% overlap)
.5 - moderate, modest, some (50% overlap)
.8 - strong, lots, a large amount (20% overlap)
Cohen’s d
Measure of effect size
What does Cohen’s d measure?
It measures how much effect the IV has on the DV
Confidence level
We are expressing our confidence in the conclusion.
This means the probability our conclusion is wrong is LESS THAN .05.
Control group…
We are not manipulating them, we are just letting them be.
Most likely given placebo
Critical Value
Value that is the deciding line between normal and deviant in a sampling distribution
Degrees of freedom…
of scores in a sample that are free to vary
df
Experimental group…
Level of the IV that receives the treatment
What is experimental method>?
Allows us to find cause and effect
Hypothesis testing
Testing whether a hypothesis is supports by the results of our research
What is inferential statistics?
Inferences are conclusions from data.
What is the interpretation?
A statement about the outcome of a statistical analysis.
Null Hypothesis…
Hypothesis that says no difference or not relationship exists between the groups being compared.
p level…
The probability we use to decide on the critical values, or cutoffs, in hypothesis testing.
Parametric test
Stat test based on assumptions about population parameters.
Quasi-Experimental Method…
IV cannot be manipulated & if we cannot randomly assign them to groups
Variable is a characteristic of the people we are studying, i.e gender.
Replication
NORMSDIST(NORMSINV(1-p)/SQRT(2))
Research Hypothesis
AKA Alternative Hypothesis
Single-group design…
measures only one group of participants
Standard error or the mean
Sx of the sampling distribution, which we can approx the stat for samples
Statistical power
Probability of being correct when we reject the null hypothesis
Statistical significance
Observed relation or difference between two descriptive measures that is unlikely to have occurred by chance.
t distribution
Set of distributions that, although symmetrical and bell-shaped, are not normally distributed.
t test
One sample t test compares a sample mean to the known population mean u.
two-tailed hypothesis
Research hypothesis in which we predict that the groups being compared are related or different, without specifying the direction.
Type I Error
Error in hypothesis testing in which we reject the null when it is true.
Type II Error
Error in hypothesis testing in which we retain the null when it is false.
If outcome is not significant.
z test
Parametric inferential statistical test of the null for a single sample where population mean & variance are known
