Chapter 15 A Flashcards
2 Principle purposes of statistics
1) to organize and summarize data–> utilizes descriptive statistics
2) justifying conclusions –> statistically test research hypothesis; utilizes inferential statistics
2 general categories of statistics
descriptive and inferential
descriptive statistics
summarizes data collected from the sample
inferential statistics
answer general questions about a population. allows to generalize results from a sample to a population
statistic vs parameter
statistics are information pieces when data is from a SAMPLE
parameter are summary values that describes a POPULATION.
frequency distribution
the number of individuals/scores in each category on the scale of measurement.
2 options for graphing numerical data (ordinal interval or ratio)
1) histogram, with BARS TOUCHING EACH OTHER
2) polygon- line graph
in a histogram/bar graph that depicts nominal or ordinal statistics, should the bars touch each other?
no. a space should be left between adjacent bars.
3 measures of central tendency
mean, median, mode
when does a mean not provide a good measure?
1) when a sample has outliers
2) in nominal scales (things that cant be quantified, like names or sex)
describing interval or ratio data
1) measures of central tendency
2) standard deviation, variance and range
3) sample variance and degrees of freesom
standard deviation
variability of the scores by measuring the distance between EACH SCORE and the mean
“average distnace from the mean”
standard deviation
variance
the average squared distance from the mean
range
the difference between largest value and smallest value in data set.
when calculating degrees of freedom, why is it necessary to use n-1?
n-1 is an adjustment when calculating sample variance in order to provide an accurate representation of its population variance.
- normal sample variance without using n-1 UNDERESTIMATES ACTUAL VARIANCE of a population
- DOF provides more accurate and unbiased view.
Empirical rule
68-95-99.7
factorial research study
includes 2 or more independent variables (ex/ temperature and humidity (IVs) on performance (DV))
how can a factorial research study be represented
via matrices
Correlation
statistical value that measures and describes the direction and degree of relationship between 2 variables ranked from -1 to +1
pearson correlation
evaluates linear relationships- if the data is interval or ratio, spearman correlation can be determined in order to measure the degree to which the relationship is ONE-DIMENSIONAL and not influenced by other factors.
regression
the equation of the straight line. (slope)
multiple regression:
when 2 variables are used to predict Y
Internal consistancy
refers to the extent to which the scores on the items correlate with each other. do the questions on the test measure the same thing?
What formula does split-half reliability use?
spearman brown formula R= (2r)/1+r
where r= correlation between the two halves of the test
R=reliability
and kuder-richardson formula #20