Six Steps of Z-Test
1. State the null and research hypothesis
2. Set the level of risk / state alpha (Usually 0.05)
3. Select appropriate test statistic (Decision rule / two-tailed)
4. Calculate test statistic / test statistic value, called the obtained value (Plug in the fo
If the null is rejected in favor of the research hypothesis:
It means that the sample average is different from the population average
If the null is not rejected:
It means that the sample is representative of the population
mean of a sample; mean of a population
The one sample z-test is used to compare the ____ to the _____.
The obtained value
Calculating the one sample z-test serves to give you:
standard error of the mean
This is the error term used as the denominator in the equation for the z value in a one-sample Z test.
One
In a one-sample Z test, this number of groups is being compared:
Accept the null hypothesis
After conducting a one-sample Z test, you arrived at a value of 0.2 for the z value. What is your conclusion?
All of the possible means
The standard error of the mean is the best estimate that we can come up with given that it is impossible to compute _______________________.
Zero
If our sample selection was perfect, what would be the difference between the sample and population averages?
2
If your sample mean is 25, your population average is 5, and your standard error of the mean is 10, what is your observed z value?
one-sample Z-test
Used to compare a sample mean to a population mean.
effect size
measure of how different two groups are from one another -- its a measure of the magnitude of the treatment
a measure of how far apart are the two scores
independent groups
Groups that are only tested once (ex/ males and females)
T-test for independent means
Looks at the difference between the average of two groups that uses a table called the t values used to determine significance of outcome
variances are equal
Which major assumption of the t test deals with the amount of variability in each group?
group one number of participants
In the formula that computes a t value, what does n1 represent?
nondirectional
In the t test for independent samples, the actual statistical test is ____
.05
In order to be 95% confident you have not committed a Type I error, at what level should you set your p value?
degrees of freedom
In order to compute the test statistic or t value, you must first approximate the sample size by calculating the __________
one-tailed test
When examining group difference where the direction of the difference is specified, which of the following is used?
95%
Under the normal curve, if the obtained value falls to the left of the critical value, what percent of the normal curve did it fall under?
t-test for dependent means
indicates that a single group of the same subjects is being studied under two conditions
One sample Z-test
Are you examining the differences between one sample and a population?
T-test for independent means
tests two distinct groups of participants, and each group is tested once
T-test for dependent means
Tests one group of participants, and each participant is tested twice
Dependent samples t-test
If you want to examine one group of subjects under two different conditions, which statistical technique should you select?
two times
The t-test for dependent means is used when your sample is tested at:
0.01
In order to be 99% confident you have not committed a Type I error, at what level should you set your p value?
...
The test statistic calculated by the statistical procedure selected is known as the ___
Obtained value
The test statistic calculated by the statistical procedure selected is known as the ___
Critical value
In order to determine whether or not you will reject the null hypothesis, the test statistic must be compared against the ___
one-tailed test
If you are hypothesizing that posttest scores will be higher than pretest scores, you should use:
53
If you are running a t-test for dependent means on a group of 54 individuals, your degrees of freedom will be:
hypothesized mean difference
Which of the following relates to the difference you expect when conducting a t-test?
post-hoc
after the fact comparisons
One-way
What type of ANOVA is used when there is only one type of treatment or grouping factor with more than two levels?
SS within
When you compute the sum of the differences between each individual score and the group mean, you have calculated the _____
N-k
When computing the degrees of freedom for ANOVA, how is the within group estimate calculated?
1
If the amount of variability due to within group differences is equal to the amount of variability due to between group differences, your F value will be equal to:
12
In a 4 x 3 factorial ANOVA design, there are how many possible group assignments for subjects are there?
Lower
If your MS within is higher than your MS between, your F value will be:
Omnibus
ANOVA first tests for an overall difference between the means. This is known as what type of test?
Type 1 error increases
If you perform multiple t-tests, which of the following is true?
True
The simple analysis of variance, or one-way analysis of variance, includes only one factor or treatment variable in the analysis.
Central Limit Theorem
Any sample distribution of a normal distribution with mean ?X and standard deviation ?X will have a sample mean of ?x and sample standard deviation of ?x /?n. Any sample distribution of sufficient size (n > 30) from any population distribution with mean ?
Cases for sampling distribution:
Case 1 - population is normal, then sampling is normal for every sample size of n
Case 2 - population is non-normal, or unknown, but sample size is at least 30, then CLT
Case 3 - populations is non-normal or unknown and sample size is less then 30, cannot
Probability
p < .05 is an example of:
T-test
used to compare samples to determine if they both come from the same population.
Z-TEST / Z-STATISTIC
used to test hypotheses about ? when the population standard deviation is known
- and population distribution is normal or sample size is large
T-TEST / T-STATISTIC
used to test hypotheses about ? when the population standard deviation is unknown
Technically, requires population distributions to be normal, but is robust with departures from normality
Sample size can be small
Thus, conditions for use:
? is unknowm
n <
degrees of freedom (df)
the number of scores in a sample that are free to vary
this limits the number of scores that are free to vary
df = n -1
-approximates the sample size
-can vary based on the test statistic selected
-
Between-subjects
Each subject receives DIFFERENT levels of the IV
Example: Child sees EITHER violent OR nonviolent cartoons
Use independent samples t-test
Within-subjects
Each subject receives ALL levels of the IV
Example: Child sees BOTH violent AND nonviolent cartoons
Use dependent samples t-test
Regression to the mean
with repeated measurements, extreme examples are likely to re-test at a value closer to the mean
Single factor designs
one IV
may use between-subjects manipulations (multiple groups receive only one level each)
or within-subjects manipulations (one group receives all levels)
Factorial designs
Multiple IVs
may use between-subjects, within-subjects, or both for the manipulation
if both are used, denoted as mixed factors design
Interactions
effects on the DV caused by on IV that change with the level of the other IVs
Main effects
effects of one IV on the DV ignoring the other IVs
Repeated measures ANOVA
ANOVA with replication is also known as?
Main effect
When analysis of data reveals a difference between levels of a factor, what is this called?
Source table
Results from an ANOVA are placed in what type of table?
Two-way analysis of variance
This is another name for the factorial ANOVA:
More than one independent variable
The factorial ANOVA is used when you have:
Main effects and interaction effects of the independent variables
The factorial ANOVA can be used to test:
The number of independent variables
The number of main effects is equal to:
a main effect
If you find that income significantly varies on the basis of gender, you have:
univariate
A factorial ANOVA is this type of analysis:
No interaction effect
When plotted on a graph, if the lines representing the effects of two variables are parallel, you have: