Unit 1 Stats

Nominal

data has to be exclusive and exhaustive

Ordinal

data can be ranked, doesn’t have equal distance between them, must be exhaustive and mutually exclusive

Interval

no zero, continuous variables, each group has exact same amount of space

Ratio

absolute zero, continuous variables, ranked categories, exhaustive, and exclusive

Population

particular group of individuals who are being studied

Sample

focus of particular study

Study elements

participants, subjects, objects that might be included in study

4 elements to a power analysis

alpha or level of significance, standard power, effect size, sample size

refusal rate

percent of subjects who decide not to participate in study

Refusal Rate equation

(# of refusing to participate / # of approaches) x 100

attrition rate

percent of students dropping out after sample size was determined

attrition rate equation

(# of dropping out / total sample size) x 100

Independent variable

variable that influences another variable and researcher controls this variable

Dependent variable

outcome varibale that is affected by another variable

Null Hypothesis

hypothesis that suggests there will be no statistically significant effect on variables being studied

Alternate Hypothesis

hypothesis that observations from a sample are influenced by a non random element

Quantitative Research

objective research used to describe, explain relationships and determine the cause and effect interactions between variables

Qualitative Research

subjective and are conducted to describe and give meaning to life experiences

Mixed method studies

uses both quantitative and qualitative methods

Probability sampling

random sampling, every member has an equal opportunity to be chosen

Simple Random Sampling

occurs through random selection of members from sampling frame

stratified random sampling

breaking into random groups and then randomly selecting from the groups

cluster sampling

randomly selecting groups

systematic sampling

starting with a number and picking every __ person

convience sampling

subjects are accesible and does not allow opportunity to control sampling errors

quota sampling

broken into known groups and selecting them

purposive sampling

consciously selects subjects to include

network/snowball sampling

use of social networks and the fact that friends have similar characteristics

Stability reliability

concerned with consistency of repeated measures of same variable/attribute with same scale or measurement over time

test-related reliability

repeated measurement of a variable over time

Equivalence reliability

compares 2 versions of same scale/instrument measuring the same event

interrater reliability

comparison of 2 observers

internal consistency / homogeneity testing

examines extent to which all items in multiple item instrument/scale consistently measure a variable

strong internal reliability

> 0.80

moderate internal reliability

0.60 to 0.80

low internal reliability

<0.60

Cronbach’s Alpha

a perfect score of 1.00 to reflect perfect reliability

Precision

degree of consistency/reliability of measurements made with physiological measure of the same variable or object under specified conditions

Validity

determines whether the measurement method accurately reflects the concept it was developed to measure

content-related validity

examines extent to which measurement contains all major elements related to concept being measured

constrict validity

determines whther the instrument actually measures the theoretical constict that is expected to measure

convergent validity

examined by comparing a newer instrument with an existing instrument

divergent validity

examined when scores from an existing instrument are correlated with scores from an instrument measuring the opposite concept

successive verification validity

achieved when am instrument is used in additional studies with a variety of subjects

criterion-related validity

strengthened when a participants score on an instrument can be used to infer performance on another variable

factor analysis

examines if an instrument included elements of the concept being measured

accuracy

involves determining closeness of agreement betwen measured value and the true value of physiological variable being measured

frequency

number of times a value for a variable occurs in a set of data

distribution

involves listing all possible values for a variable in a study

percentile ranks

can be used to analyze ordinal date that can be assigned to categories that can be ranked

measure of central tendency

used to describe study variables

mode

score that occurs with greatest frequency in a distribution of scores

bimodal

2 scores are the most frequent

multimodel

more than 2 modes

what is mode best for

nominal

median

score that lies in the middle

what is median best for

ordinal

range

distance between smallest and greatest values

mean

average of all scores in a sample

what is mean best for

ratio and interval

outlier

value that is unusual and skews data

line graphs

joined number of series of plotted points with a line to illustrated how a variable changes over time with descriptive data and patterns in the data

what does the x-axis document on a line graph

time

what does the y-axis document on a line graph

mean scores

scatterplots

describes relationships between 2 variables to provide a graphic representation of data from a study

positive relationship

points move from lower left to upper right

negative relationship

points move from upper left to lower right

weak relationship

<0.3 or > -0.3

moderate relationship

0.3 to 0.5 or -0.3 to -0.5

strong relationship

>0.5 or < -0.5