Basic Terms and Definitions Flashcards
Descriptive statistics
summarize, understand, and describe a group of numbers from a research study
What are examples of descriptive stats
- measure of central tendency (mean, median, mode)
2. variability (range, varience, sd)
Inferential statistics
- draw conclusions and make inferences based on numbers from a research study, but go beyond these numbers
- if there is a difference between groups
- how sure we are about our conclusions
Variable
condition or characteristic that can have different (varying) values (ex. male/female, …22,23,24…,
Score
the value of a particular person’s answer (ex. male, 25yrs old, 15 sexual partners)
Variability
how much change there is in a group of scores
Variance
a specific measure of variability
Correlational Research Designs
examines if there is a relationship between variables
- “what does knowledge of x tell us about y”
- naturally occurring variables
Experimental Research Designs
goal is to determine why something happens
- “what causes X?”
- control and manipulation of variables
Independent variable
variable that the researcher manipulates/changes
Dependent variable
variable that changes because of IV
Components of an experiment
- Must have more than one level of IV
- several, stable/reliable DVs
- control variables
- random assignment
Control variables
keeping everything else constant
Random assignment
purpose of random assignment is to create equivalent groups
Population
set of all cases of interest, generally a theoretical concept (not always measurable)
Sample
subset of population that is being studied; something to be concerned about is biased samples
Biased Samples
sample that systematically underselects or overselects from certain groups in the population
Parameter
Some characteristic of population
Statistic
some characteristic of sample
Scales of measurement
- Nominal
- Ordinal
- Interval
- Ratio
Nominal
mutually exclusive categories (yes or no)
Ordinal
placing variables in a “ranked” order (no, somewhat, yes)
Interval
equal distances between points on the scale (temp., on scale from 1-10 how helpful)
Ratio
exactly like interval scale, but has a true zero (#oof cigs smoked today)
IV measurement
often at a nominal level
DV measurement
Data analysis is limited when using nominal or ordinal data
- interval data is desirable
Unimodal data
only one very high area (mode)
Bimodal data
two very high areas
Multimodal data
many very high areas
Rectangular data
all values have about the same frequency
Symmetrical data
many psychological variables (height, weight, attitudes, work productivity); distribution looks the same on both sides if you split it down the middle
Skewed data
majority of responses are at low or high ends of the scale
Central tendency
a measure that refers to the typical score in a distribution
- “center of data,” “best representation”
Measures of central tendency
mean, median, mode
mean
average - interval/ratio data
median
middle score - ordinal data (very skewed interval/ratio data)
mode
most frequent - nominal data - bimodal distribution
Properties of the mean
- the value is the value around which all observed values are balanced
- the point at which the sum of the squared deviations is minimized
- influenced by extreme values (main disadvantage)
Variability
the spread of scores in a distribution
Types of variability
range, variance, standard deviation
Range
the distance between the larges and smallest scores (only reflects two most extreme scores
Variance
the sum of the squared deviations from the mean, divided by N-1
Steps for calculating the varience
- subtract the mean from each of the scores to get deviations
- square each deviation
- add all squared deviations
- divide by the total number of scores minus 1
Standard deviation
the most widely used measure of variability (square root of variance)
