quantitative data analysis Flashcards
descriptive statistics
are summary statistics that allow the researcher to organize data in different ways that give meaning and facilitate insight
can be used to describe the sample
ex mean age, education level, gender
inferential statistics
statistics designed to allow inference from a sample statistic to a population parameter
- allows the researcher to estimate how reliably they can make predictions and generalize findings
what are the 4 levels of measurement
nominal, ordinal, interval, ratio
nominal
- classified in mutually exclusive categories
- no ranking within the categories
ex: gender, marital status - mean, mode, frequency of distribution
ordinal
- data must be mutually exclusive and exhaustive and is sorted on the relative ranking of variables
example: education level or a likert scale - mode, median
- rank order of coefficients, range, percentile
interval
mutually exclusive categories, exhaustive categories and ranking order plus the distances between the intervals are numerically equal
- no zero point on the interval scale
ex: temperature
- mean, median and mode
- range, percentile, standard deviation
ratio
- highest level of measurement
- mutually exclusive, exclusive and exhaustive categories, ranking order, equal space between intervals and a continuum of values
ex: weight, length and volume - absolute zero exists - can be an absence of weight
- mean, median and mode
- range, percentile, standard deviation
frequency distribution
the number of times each event occurs is counted and data is then grouped according to categories; the frequency of each group is reported
mean
average
calculated by summing values and dividing that sum by the number of values
median
the midpoint in a set of values
50% of distribution fall below the median and 50% above the median
mode
the most frequently occurring score in the distribution
normal distribution
a theoretical concept that observes that interval or ratio data group themselves about a midpoint in a distribution closely approximating the normal curve
- mean, median and mode are equal
positive skew
low range mean
negative skew
high range mean
range
is the difference between the highest and lowest scores
simplest but most unstable measure of variability
percentile
is the percentage of cases a give score exceeds
- median is the 50th percentile
- a score in the 90th percentile is only exceeded by only 10% of scores
standard deviation
the average amount of variability in a set of scores or the scores average deviation from the mean
- a measure of how dispersed the data is in relation to the mean, calculated using statistical formula
- the average distance from the mean
- low standard deviation = data are clustered around the mean
- a high standard deviation = data are more spread out
inferential statistics
combine mathematical processes with logic and allow researchers to test hypotheses about populations by using data obtained from probability samples
purpose:
- to estimate the probability that statistics found in the sample accurately reflect the population parameter
- test a hypothesis about a population
parameter
a characteristics of a population
- a well defined set that has specific properties
statistic
is a characteristic of a sample and is used to estimate population parameters
parametric tests
are statistical procedures that can be used when three assumptions are present
- the sample from the population has a normal distribution
- level of measurement must be interval or rate with a normal distribution
- sample was obtained through a random sampling procedure
non-parametric tests
are statistical procedures that can be used when
- the sample from the population does not have a normal distribution
- level of measurement is nominal or ordinal
- sample was obtained through a non-random sampling procedure
hypothesis
H1
a formal statement of the expected relationship between 2 or more variables in a specified population
null hypothesis
H0
states there is no relationship between the variables being studies, used for testing and interpreting statistical outcomes
type I error
rejection of the null hypothesis when it is actually supposed to be retained
- stating a relationship exists when it does not
type II error
retaining the null hypothesis when it should be rejected
- can occur if sample is too small
- stating there is no relationship when there is one
level of significance or alpha
is the probability of making a type I error
- 0.05
statistical significant hypothesis
unlikely that the findings have occurred by chance
- if the alpha is 0.05, then 95% change the researcher will make the correct conclusion
practical significance
is the practical value that the study contributes