Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
99 Cards in this Set
- Front
- Back
probability samples
|
aka random samples
each member of a population has an equal likelihood of being selected |
|
dependent variable (dv)
|
variable we MEASURE (ie: data)
think of DV as EFFECT in terms of cause & effect predicted variable (y) |
|
independent variable (iv)
|
variable we MANIPULATE
think tx administered predictor variable (x) |
|
results of research are always _______? why?
|
tentative!
data never ends; need replication |
|
what makes experimental research TRUE?
|
meets 3 criteria
1. use a control group 2. manipulate IV 3. randomization |
|
what is Quasi-experimental research?
|
biased samples
there is lack of randomization in selection of samples |
|
other names for quasi experimental?
|
volunteer
non probability biased |
|
what is significant about conclusions in biased/quasi experiments?
|
cannot generalize to (N) population
results are SAMPLE SPECIFIC |
|
what is a confounding variable?
|
a variable highly related to IV &/or DV that may effect accuracy of results
|
|
what do parameters refer to?
|
the characteristics of population
|
|
population uses ______ letters
|
greek
|
|
MU refers to what?
|
population mean
|
|
Sigma2 refers to what?
|
population variance
|
|
Sigma refers to what?
|
population standard deviation
|
|
RHO refers to what?
|
population measure of bivariate relationships
|
|
what is the funny looking "P" symbol?
|
greek letter RHO = population measure of bivariate relationships
|
|
what refers to the characteristics of the sample?
|
statistics
|
|
(n) sample uses ______ letters.
|
roman/latin/english
|
|
________ estimates the value of ________
|
STATISTICS estimate the power of a PARAMETER
|
|
_
X what does this symbol signify? |
sample mean
|
|
S2 refers to what?
|
sample variance
|
|
S refers to what?
|
sample standard deviation
|
|
r refers to what?
|
sample measure of bivariate relationships
|
|
what is empirical research?
|
scientific research we observe and measure
|
|
whats a research design/method?
|
a plan for collecting and analyzing data that allows us to answer specific empirical questions
|
|
define conceptual definition
|
any definition you'd find in a dictionary
|
|
generalization has to do with what?
|
taking a sample (n) of a population (N) and generalizing the results back to the population
|
|
operationalization has to do with what?
|
part of the research design which is used to answer empirical (scientific) questions
|
|
define operational definitions.
|
defines conceptual definitions (subjective defs) in terms of how they are measured
ie: scale |
|
what is the funny looking E?
|
uppercase sigma
signifies summations "add em up" |
|
the act of assigning numbers to observations according to a set of rules is known as what?
|
measurement
|
|
what are the rules of measurement
|
1.nominal scale
2.ordinal scale 3.interval scale 4.ratio scale |
|
what is a nominal scale?
|
numbers represent categories or classifications
numbers are used to identify no arithmetic can be performed (ie: SS#, DL#, Cell #, home address) |
|
what is an ordinal scale?
|
rank order, standings, percentile rank
unequal distances between points on a scale |
|
whats an arbitrary zero? what scale does it have to do with?
|
means no true zero point; does NOT signify absence but a point on a scale
think Fahrenheit scale; 0 is not absence of temperature interval scale |
|
what is an interval scale?
|
equal distances between points on a scale
no true zero point (arbitrary zero) |
|
what is a ratio scale?
|
equal distances between points on a scale
TRUE ZERO POINT (ie: odometer, bank account) |
|
what is reliability?
|
CONSISTENCY of measurement
|
|
is the measured instrument measuring consistently?
|
reliability
|
|
whats is validity?
|
ACCURACY of measurement
|
|
is the measured instrument measuring what its supposed to measure?
|
validity
|
|
what is the mode?
|
most frequently occurring score
|
|
what is the median?
|
middle score
|
|
what is the mean?
|
arithmetic average
_ X= sumX/n |
|
describe a normal distribution
|
mean, median, and mode are all the same
|
|
what are some names for normal distribution
|
gaussian distribution
bell curve symmetrical distribution |
|
how can extreme outliers influence the mean?
|
can mislead the measure of central tendency
large numbers pull mean high |
|
tell me about positive skew
|
-there are many small numbers
-numbers are bunched in lower end of distribution -tail points in positive direction |
|
how does a positive skew affect mode, mean, and median?
|
mean is larger than median is larger than mode
|
|
tell me about negative skew
|
-many large numbers
-numbers are bunched in upper end of distribution -tail points in negative direction |
|
how does a negative skew affect the mean, median, and mode?
|
mean is smaller than median is smaller than mode
|
|
what are measures of dispersion?
|
measures the degree to which scores are spread out from eachother
|
|
what are 3examples of dispersion?
|
range
variance standard deviation |
|
what are 3examples of central tendency?
|
mean
median mode |
|
how do you find the sum of deviation scores?
|
find the mean in the group of datapoints.
__ subtract mean from x to get (x-X) sum will always equal zero! |
|
why do you square the sum of deviation scores?
|
the sum of deviation scores always equal zero so you need to square them to get a number to work with
|
|
what are standard normal distributions used for?
|
scientists use them to describe many events that happen in nature
|
|
whats standardization?
|
set of procedures in which raw scores are transformed to another scale with a known mean and standard deviation for the purpose of determining percentile rank
|
|
what process do you use to find percentile rank?
|
standardization
|
|
what is the distance from the mean to any score in the distribution in standard deviation units
|
Z score
|
|
what are the details about Z score?
|
ranges from -3 to 3 with a mean of zero and a standard deviation of 1
|
|
when determining percentile ranks what are the rules?
|
1. if Z is positive add 50% (.5000) to area between mean and Z
2. if Z is negative we simply use an area beyond Z |
|
why do we use Z scores?
|
to compare many different measures that have different means and standard deviations, it converts them to a common scale
|
|
"statistics" ... is what?
|
the language of science
|
|
correlation relationships are..
|
linear
|
|
the extent to which 2variables change in a systematic way or COVARY is known as..
|
correlation
|
|
if correlations are high enough it gives us the ability to predict y from x using...
|
simple linear regression
|
|
y1=bx + a
explain the symbols |
y= predicted variable (DV)
1=prime b=slope x= predictor variable (IV) a= y intercept |
|
"statistics" ... is what?
|
the language of science
|
|
correlation relationships are..
|
linear
|
|
the extent to which 2variables change in a systematic way or COVARY is known as..
|
correlation
|
|
if correlations are high enough it gives us the ability to predict y from x using...
|
simple linear regression
|
|
y1=bx + a
explain the symbols |
y= predicted variable (DV)
1=prime b=slope x= predictor variable (IV) a= y intercept |
|
correlations do not infer..
|
causation
|
|
as correlations ___, prediction error _____
|
corr increase, error decrease
corr decrease, error increase |
|
y1=bx+a
formula is for what |
plotting the line of regression
aka line of best prediction, line of best fit |
|
if correlations are high enough it gives us the ability to predict y from x using...
|
simple linear regression
|
|
what does a positive correlation scatterplot look like?
|
points up to the right
points down to the left eliptical shape as x inc, y inc as y dec, x dec |
|
what does a negative correlation scatterplot look like?
|
points up to the left
points down to the right eliptical shape as x inc, y dec as y inc, x dec |
|
SestY=_____
what is this |
standard error of the estimate the prediction error formula
|
|
which is the predicted /predictor?
x & y |
y=predicted
x=predictor |
|
what is r2 (r squared)
|
coefficent of determination
|
|
the coefficent of determination is to measure what?
|
amt of variation x&y have in common
amt of shared variation |
|
if r2 is .76 what does that mean?
|
that 76% of the variation in y can be explained or predicted in the variation of x
|
|
what is
1-r2 (r squared) |
coefficient of non determination
|
|
what does the coefficient of NON determination measure?
|
the amt of variation x&y DONT have in common
amt of UNshared variation |
|
what does this mean?
1-r2=.24 |
24% of the variation cannot be explained or predicted by the variation of x
|
|
what scales apply to the pearson r
|
ration & interval
|
|
what scales refer to spearmans rho?
|
ordinal
|
|
what scales refer to point biserial correlation coefficient aka phi coeff
|
nominal
|
|
what is a spurious correlation?
|
correlation that artifically inc or dec bc of some confounding factor
|
|
what is curvilinearity
|
rainbow shaped scatterplot
pearson r is meaningless |
|
the value of the correlation coefficient always lies in what range?
|
-1 to 1
|
|
tell me about correlation strength
|
ranges -1 to 1
to closer to 1 or -1 the closer the points are in scatterplot the closer to 0 the weaker the line and more spread out the points 0=no linear correlation just randomly spread out points |
|
what does r=1 look like on a scatterplot
|
r=1 is perfect positive linear correlation; pts lie in straight line pointing up to right
|
|
what is descriptive statistics
|
methods for organizing, displaying, and describing data by using tables, graphs, and summary measures
|
|
what is inferential statistics
|
methods that use sample results to help make decisions or predictions about a population
|
|
what is a discrete variable
|
values are countable
(ie: number cars sold) |
|
what is a continuous variable
|
variable that can assume any numerical value over a certain interval
ie: exam is between 30-60min |