Stats
what is the name of the statistic that pertains to area "B" when none of the variables in the equation are standardized?
Beta
An effect size that is typically used when one variable is dichotomous
D
summary of regression results
In regression analysis, _________ was predicted using _________,_________ and __________. The three predictors explained a _____ portion of variance in ___________ (R^2 = ______, *F(p,n)*=______, p<______). Correlations and standardized betas are reported in Table ______. If correlation and SB are sig - There was a (small/medium/large) (positive/negative) correlation between ______________ and ___________, and _______________ explained unique variance in ____________ after controlling for the effects of the other variables in the equation. if just correlation is sig - although there was a (S/M/L) (p/n) correlation between ____________ and ________, the effect became non-significant after controlling for other variables. if neither is sig - Notably, _________ was not significantly related to ______________.
Two distributions of sampling error that are skewed in shape
L & X^2
The assumption that the population parameter for a statistic is equal to zero
Null hypothesis
The probability of making a correct decision if the null hypothesis is false
Power
An effect size that is typically used with two numerically scaled variables
R
what is the name of the statistic that is equal to srt(B+C+D)/(A+B+C+D)
R
what is the name of the statistic that is equal to (B+C+D)/(A+B+C+D)
R^2
three biased statistics
R^2, R, r^2
A word that means "if the null hypothesis is true, then the results from the sample are extremely unusual"
Significant
what is the name of the statistic that pertains to area "B" when all of the variables are standardized prior to analysis?
Standardized beta
A distribution of sampling error that is almost normal in shape
T
Written summary for cronbach's alpha
The satisfaction has a _______ reliability (cronbach's alpha = ____) these results suggest that ___% of the variance was measurement error and that on average, a participants observed score *is likely to deviate* from their true score by _____ unites
summary of post hoc comparisons
There was a significant difference between groups (n^2 = ____, F(p-1, n) = ______, p =_____). Means and standard deviations for each group are reported in table 1. Participants in teh control group were significantly more/less _____ than people in the ________ group (d = _______). This represents a small/medium/large effect. Control couples were also significantly more/less ________ than people in _______ group (d = _____), and this represents a ______ effect.
A persons average score on a test if he were assessed an infinite number of times
True score
A regression parameter estimate for which a value of .8 could indicate a smaller effect than a value of .2
Unstandardized beta
What value is X(hat) equal to?
X (bar)
what are two things that increase the size of the standard error?
a large standard deviation, a small sample
equation for intercept
b0 = Y(bar) - b1*X(bar)
equation for standardized beta
b1* (sd(x)/sd(y))
equation for t value for the beta weight
b1/sderror
Bruno computed a standard deviation, but he accidentally used "n" rather than the degrees of freedom in his formula. This means that the value he obtains will be ____________________ .
bias
The term used when the expected value is not equal to the parameter.
bias
a high correlation between two predictor variables in a regression equation
collinearity
A type of research that involves observing that which occurs naturally without manipulation
correlational research
equation for beta weight
cov/var(x)
a statistic that indicates how much additional variance is explained when one set of predictors is added to another set
delta R^2
A numerical value based on data from a sample that indicates something about the sample.
descriptive statistic
what is a paired sample correlation for?
determine if the rank order of cases show stability across different assessment points.
`standard error of the mean reflects
deviations around a population parameter
what are two assumptions of an independent samples t-test?
equality of variances and independence of observation
The average of all the values for a population of values from which a single value can be randomly sampled.
expected value
A type of research that involves manipulation and random assignment.
experimental research
what is the standard error of measure?
if you test me many times my scores will change and stray from true. the standard error of measure says just how far they do
A numerical value based on data from a sample that indicates something about a population.
interential statistic
types of sampling error
mean, mean diff, correlation, beta
standard error of measure pertains to
observation of a single case
a numerical value based on data from all the cases in a population that describes something about the population
parameter
A parameter estimate in multiple regression that is a correlation
partial correlation
the resulting beta when all the variables are residuals that have been turned into Z scores
partial correlation
what is the name of the statistic that is equal to srt(B+C)/(A+B+C+D)
r
what is the name of the statistic that is equal to srt(B)/(A+B)
r partial
what is the name of the statistic that is equal to srt(B)/(A+B+C+D)
r semipartial
what is the name of the statistic that is equal to (B+C)/(A+B+C+D)
r^2
what is the name of the statistic that is equal to B/(A+B)?
r^2 partial
what is the name of the statistic that is equal to (B)/(A+B+C+D)
r^2 semipartial
The set of cases from which a researcher collects data.
sample
a regression parameter estimate that can be interpreted as having a scale that is similar to a correlation
standardized beta
the resulting beta when all the variables are turned into Z-scores prior to analysis
standardized beta
when a bivariate correlation between two variables is smaller than the partial correlation between those variables
suppressor effect
summary of regression results w/ multiple steps
the the first step of a regression analysis, _________ was predicted using ______ and ________. These two preictors explained a significant portion of variance in ___________ (R^2 = ___, F(p,n) = _____, p < .001). If correlation and SB are sig - There was a (small/medium/large) (positive/negative) correlation between ______________ and ___________, and _______________ explained unique variance in ____________ after controlling for the effects of the other variables in the equation. if just correlation is sig - although there was a (S/M/L) (p/n) correlation between ____________ and ________, the effect became non-significant after controlling for other variables. if neither is sig - Notably, _________ was not significantly related to ______________. in the second step, *ratings of ________, ________, and _____* were added to the equation. These variables explianed a significant portion of additional variance in _________ (delta R^2 = ______, delta F (p,n) = _______, p< _____), and *all 5 variables taken together explained ______% of the variance*. Correlations and standardized betas are reported in Table 3. If correlation and SB are sig - There was a (small/medium/large) (positive/negative) correlation between ______________ and ___________, and _______________ explained unique variance in ____________ after controlling for the effects of the other variables in the equation. if just correlation is sig - although there was a (S/M/L) (p/n) correlation between ____________ and ________, the effect became non-significant after controlling for other variables. if neither is sig - Notably, _________ was not significantly related to ______________.
written summary for independent T-test
there was a significant difference between _________ and _____ (D=___, t(n,n)=______, p<_______) with _______ reporting more/less _____ (M=____, SD=______) than _______ (M=______,SD=_________). This is a _____ effect.
The extent to which a scale measures what it is supposed to measure.
validity
what is significance?
when p is less than 0.05
In the process of computing a standard deviation, Zelda lost one degree of freedom. What did Zelda do that caused her to lose one degree of freedom?
when we calculate the mean we lose one degree of freedom because it means that one variable cannot vary
equation for predicted score when r=0
y (hat) = y (bar)
equation for residual scores
y - y(hat)
equation for predicted score when r=1
y(hat) = y
3 equations to calculate b1
y=b0 + b2x2 +e(y.x2) x1 = b0 + b2x2 +e(x1.x2) e(y.x2) = b0 + b1e(x1.x2) + e
What is a limitation of correlational research?
you can't determine causes
What is a limitation of experimental research?
you cant explain natural phenomena
cutoffs for correlation
< .1 = trivial > .1 = small > .3 = medium > .5 = large
cut offs for d-statistic
< .2 = trivial > .2 = small > .5 = medium > .8 = large
cut offs for cronbach's alpha
.8 = good .7 = adequate < .7 = poor
This statistic describes a distribution that becomes almost perfectly normal in shape as the sample size increases beyond 30
Sd x
This will become smaller as the sample size increases
Sd x
This is equal to true variance
Sd^2 - sd^2 measure
This is estimated by a correlation between two alternative forms
Sd^2 true / sd^2
This is estimated by a test retest correlation
Sd^2 true / sd^2
In class, we discussed two key questions that could be addressed if Moniz had conducted a clinical trial. What are those questions?
how big is the effect? and could it be random chance?
what is the standard error of a statistic?
how much *change* do you typically find if a study is repeated many times? each statistic we declare will have its own standard error
what is a d-statistic?
it says how far 2 means are split, measured with sd
standard deviation pertains to
observations of different cases
standard error of estimate pertains to
observations of different cases
standard error of the correlation pertains to
observations of different samples
standard error of the mean difference pertains to
observations of different samples
standard error of the mean pertains to
observations of different samples
All of the cases in which a researcher is interested.
population
equation for mean square regression
regression sum of squares / p
equation for R^2
regression sum of squares/ total sum of squares
The extent to which a measured value stays the same regardless of how, when, or where it is measured.
reliability
sd of residual scores when r=o
sd of y
equation for total sum of squares
(sd^2)(df)
The probability of committing a type 1 error if the null hypothesis is true
Alpha
The probability of committing a type 2 error if the null hypothesis is fals
Beta
What are the lyrics for the mean?
If you pick someone at random and you want to bet, the mean is the score you expect to get
The standard error of measure.....
Indicates the typical deviations for eai from the linear equation Xai = truei + eai where the subscripts A and I stand for *assessment occasion* and individual
A standard deviation...
Indicates the typical deviations for eis from the linear equation Xis= Xs + eis where the subscripts i and S stand for individual and sample
A standard error of estimate.....
Indicates the typical deviations for eis from the linear equation Yis = (predicted, based on Xis) + eis where the subscripts i and S stand for individual and sample
The standard error of a mean difference .....
Indicates the typical deviations for esp from the linear equation (y-y)sp = (m-m)p +esp where the subscripts s and p stand for sample and population
A standard error of the correlation....
Indicates the typical deviations for esp from the linear equation Rsp = Pp + esp where the subscripts s and p stand for sample and population
Written summary if z- score
Score was ___ sd below the mean. If the score is normally distributed, the score is between ____th and ____th percentile
This is the standard error of measure
Sd (srt 1 - reliability)
This is estimated by cronvach's alpha
Sd ^2 true / sd^2
Written summary of mean and sd
On a scale that ranged from ____ to _____ the average score was ____. On average, people deviated from this score by _____ units
This value will become smaller as the effect size increases
Sd est
what is the difference between a variable and a parameter estimate?
The difference between a variable and a parameter is that variable describe a sample. A parameter describes an entire population.
What is xi?
The score on x for case i
write the two linear equations associated with a paired sample estimate for mean difference score
Xdiff is = X(bar)diff s + ediff is X(bar)diff sp = mdiff p +ediff sp
write the two linear equations associated with a single sample estimate for a single mean
Xis = X(bar)s + eis Xsp = mp +esp
The probability of committing a type 1 error if the null hypothesis is false
Zero
what is t-value?
a statistic divided by its standard error
whats a point biserial correlation for?
determine if the t-test for mean differences was significant, determine if there was a difference between means
what is the levene's test for?
determine if there is a difference between standard deviations, and determine if any assumptions are violated
standard error of the correlation reflects
deviations around a population parameter
standard error of the mean difference reflects
deviations around a population parameter
standard error of estimate reflects
deviations around a predicted value
standard deviation reflects
deviations around a sample mean
standard error of measure reflects
deviations around a true score
what are two different and unrelated types of effects that can be tested in a paired samples analysis?
difference between means, correlation between pairs
Bert and Ernie each conducted a research study and they they both used the Beck Depression Inventory as an outcome variable. Although the mean and standard deviation was the same in both studies, the standard error of the mean was larger in Bert's study. What did Bert do that caused the standard error to be larger in his study?
smaller sample size
Kanga conducted a study and every participant completed two different measures of deprrestion. The standard error of one was larger than the other. What caused one to be larger?
smaller sample size
equation for standard error of estimate using an ANOVA table
sqrt (residual sum of squares /n-p-1)
Write the formula for a standard deviation.
square root of (Xi-X)^2/n-1)
What commonly calculated statistic is essentially the mean of |ei| ?
standard deviation
The general term for a numerical value based on data from a sample (that may indicate something about the sample or something about the population).
statistic
What are the lyrics for the standard deviation?
the standard deviation is used to measure things, its how far the average person strays form the mean
what is the standard error estimate?
the standard error of estimate, tell me what its for. Got people in a sample and you guess their score. Its your typical mistake, no less no more
equation for variance of a dependent variable
total sum of squares / n-1
equation for the sum of squares for the explained variance
y(hat)- y (bar)
equation for the sum of squares for the error variance
y-y(hat)
write the three equations associated with an independent sample test for a difference between two means
yis =y(bar)0s +(y(bar)1s - y(bar)0s)xis + eis y(bar)0sp = m0p + e0sp (y(bar)1sp - y(bar)0sp) = (m1p - m0p) e1-0sp
