exam 2 (ch 6-8) Flashcards
two variables to have a distribution
bivariate data
when scores on 2 variables are paired
bivariate data
a quantitative relationship between two variables
correlation
correlation does or does not mean causation?
not
what kind of correlation is this: drunkness and motor skills?
negative correlation
what kind of correlation is this: coffee intake and energy levels?
positive correlation
what kind of correlation is this: hours on videos and length of toes
no correlation
two fundamental techniques when it comes to stats
correlation and regression
a stat that represents the strength and direction of a relationship between two variables
correlation coefficient
correlation coefficient: the value of zero=
no relationship
correlation coefiencent: value near -1 or 1 indicates
strong relationship
correlation coefficient: values near zero but not zero
weak relationship
use to illustrate a correlation
scatterplot
the population of variance that two variable share
coefficient of determination
coefficient of determination is also known as
shared variance or common variance
a stat technique that allows you to make predictions
linear regression
correlation is used to establish predictions or relationships?
relationship
linear regression operates on what?
equation of a line (y=mx+b)
distributions can be broadly categorized as what two things?
theoretical or empirical
collecting data and plotting them on a frequency histogram
empirical distribution
the distribution of scores from actual data
empirical distribution
the distribution of scores one would expect to find
theoretical distribution
hypothesized via math logic/ probability
theoretical distribution
three types of theoretical distribution
normal, binomial, and rectangular distribution
a measure of the likelihood that an event will occur
probability
probability is at the heart of
theoretical distribution
distribution in which all scores have the same frequency
rectangular distribution
equal probability; all scores have the same probability of occurring
rectangular distribution
distribution of the frequency of events that can have only two possible outcomes
binomial distribution
bell-shaped theoretical distribution that predicts the frequency of occurrence of chance or random events
normal distribution
standard deviation units are used to express measurements for what type of scores
z scores
each trial is asymptotic which means
it never touches x axis; it goes on indefinitely
A point that separates concave upward or concaves downward
inflection point
a subset of a specified group
samples
we use samples because
we do not have access to the population usually
characteristics of a sample
stats
entirely of a specified group
population
characteristics of a population
parameter
a subset of a population chosen so that each member (item/data point) has an equal probability of being selected
random sample
best method of sampling
random sample
a subset of the population chosen such that not all members (items.data points) have an equal chance of being selected
bias sample
the degree to which your sample is reflective of the population you’re interested in
representativeness
representativeness can be achieved one of the two ways
via sampling technique and via large sample size
based on all possible random samples drawn from the same population (using theory of probability)
same distribution
the mean of sampling distribution
expected value
standard deviation of a sampling distribution
standard error
t/f: every sample of a sampling distribution of the mean is carefully selected from the same population size
false, it is random
t/f: sample size is the same for all samples
true
t/f: the number of samples is very small, less than 20
false
t/f: the mean (x bar) is calculated for every sample
true
the sample means (all the x bar) are arranged into what?
a frequency distribution
as the sample size increases, the sample distribution looks more like the theoretical normal curve
central limit theorem
the expected value (x bar of sampling distributions) is equal to
standard error
t/f: the standard deviation of the sampling distribution is not equal to the standard deviation of the population
true
if we dont have access to the parameters what can we use?
t distribution table
any continuous probability distribution that arises when estimating the mean of a normally distributed population in situations where the sample size is small and/or the population standard deviation is unknown
t distribution
determines the distribution that is most appropriate for your sample size
degrees of freedom
you need three things to read a t -table
- a degrees of freedom
- determine if it’s 1 or 2 tail
- determine the alpha/confidence level
the probability you are willing to be correct/incorrect about something
alpha level/confidence
range of scores that are expected to contain a parameter (i.e. population mean)
confidence intervals
