Reading...
Play button
Play button
Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
80 Cards in this Set
- Front
- Back
|
Mode (NOIR)
|
Category or score that occurs the most
NOIR- Only one in which you can use Nominal level variables Highest # of repeating data Eg. (122579) the mode would be (2) (12233444567) mode=(3,4 are bi-model values) |
|
|
Median (OIRI)
|
Divides an array of values into two equal halves, midpoint
least effected by outliers of all three, good to use with outliers Eg. (1223456789) Median 4.5-Even amounts of numbers (345678910) Median would be 6-odd numbers |
|
|
Mean
|
Ratio- the average of values,highest level of measurement
best represents the central tendency use w/ largest # of cases, even w/ outliers good with large # of values Add ll values in data set & divide by the total number of values in set Eg. (3,4,5,6,7,10,11) 46/7=6.57 |
|
|
What are the 4 levels of measurement?
|
Nominal
Ordinal Interval Ratio |
|
|
Nominal
|
Categorizes values into descrete subclasses
Must have 2 or more values No numerical value Uses value categories Only use w/mode Eg, Do you believe govt. should provide health care? 1. yes 2. no 3. undecided |
|
|
Ordinal
|
Numerical values, preserves rank order
Can rank order values from high to low and most to least Use w/median Does not indicate absolute quantities or assume equal intervals btwn. categories Eg. How would you rate your social worker? 1 very good 2. good etc. |
|
|
Interval
|
Can use to rank order diff. measurements of a variable
Places the value for the variable on an equally spaced continuum Preserves rank order |
|
|
Ratio
|
preserves rank order, unit differences and fixed zero points
|
|
|
Skewed Distribution
|
Symmetrical, ends do not tapper off
Positively skewed= tail curves to the right Negatively skewed= tail curves to the left |
|
|
Normal distributions
|
Symmetrical bell shaped curve that often arises when a trait is composed of a large number of random independent factors; the curve possesses a specific mathematical formula
Interval & Ratio Level |
|
|
Normal Curve
|
Interval or ratio level variable that is normally distributed represented by a bell shape curve
The Mode,Median & Mean all occur at the highest point in the center of the distribution |
|
|
Graph types
|
bar
line pie histogram Pareto frequency polygons stem & leaf |
|
|
Graphs
|
can communicate the bigger picture.
Improve communication Is possible to lie w/graphs like other analysis |
|
|
Bar Graph
NOIR |
NOIR
The highest of each bar reflects the frequency of the value category Bars of equal width, do not touch each other |
|
|
Line
|
Flip the access & put the frequency on the bottom
|
|
|
Pie-Charts
NOIR |
NOIR used w/all levels
use "absolute percents" when making pie chart |
|
|
Histogram
OIR |
Looks like bar, but the bars touch each other
Uses height of bar to reflect the frequency of a value or value category Bars can be various widths, however, if the are "equal" must be Ordinal |
|
|
Pareto
IR |
Portrays rank ordering of values of a variable by frequency in descending order.
Reflects "cumulative frequencies & cumulative percentages" of cases at any point or when added to the group |
|
|
Frequency Polygon
IR |
picks points to connect to
|
|
|
Stem & Leaf
IR |
Displays all of the actual case values in he distribution of a variable
Stem= 1st or 2nd digits of a persons age Leaf= 2nd digit of a persons age |
|
|
Which graphs are used more often?
|
Frequency Polygon & Histogram
|
|
|
Central Tendencies
|
Shows what is typical with the data
|
|
|
Trimmed Mean
|
Designed to minimize the effect of a few extreme outliers.
Combines the best features of Mean & Median Usualy top 5% & the bottom 5% of values in an array are thrown out. The remaining 90% of values are averaged. *This is trimmed mean* Use w/small amount of values & outliers **Less than 30 points of data use trimmed mean |
|
|
Weighted Mean
|
The weighing fo numbers in order to arrive at a value that is more meaningful for the data set than either the arithmetic mean or trimmed mean.
A. Summarized data B. Provide a common reference point to compare 2 groups of data |
|
|
Correlation
|
The degree of association btwn. or among variables, figuring out trends effectiveness, some may be weak & some may be strong or moderate (does not predict what other variables will do)
|
|
|
Covariance
|
The degree to which certain values of one variable are found to be associated with certain values of other variables
|
|
|
Scattergram
|
Displays the correlation btwn. 2 or more variables, it simultaneously portrays the info. for 2 values- not more than 2 values. If you see a pattern form it could mean that a relationship exists
|
|
|
What is the highest correlation?
|
+1.0
|
|
|
What is the weakest correlation?
|
-1.0
|
|
|
What is NO correction?
|
0 (zero)
|
|
|
How do you determine if a relationship exists?
|
-1.0 0 +1.0 (these are the numbers you look at to determine if a relationship exists, if 2.5 is your #, it is wrong because you can't go over 1.0 or -1.0
|
|
|
1 Tail Test
|
Predicts relationship btwn. variables, predicts a specific direction
Eg. Watching hockey increases beer consumption (A relationship btwn. the two, it takes it sin some sort of direction either improves, or increases or diminishes etc.) |
|
|
2 Tail
|
Predicts relationship, but no specific direction
Eg. Gender is related to job satisfaction levels (does not predict whether M or F will be found to have higher levels of job satisfaction) Proves some sort of relationship btwn. two variables, does not increase or decrease |
|
|
Null Test
|
x=y, says there is no difference btwn. the population & the sample population
|
|
|
Dichotomous Variable
|
Has 2 value categories
Eg. Gender, T/F |
|
|
Binary Variable
|
Assigns numerical value to categories of 1 or 0 to indicate presence, or absence of variable. Used when coding
|
|
|
Dummy Variables
|
Created by converting a qualitative into a binary variable, breaks it into sub-sets.
Eg. gender=make & female |
|
|
Information
|
analyzed data
|
|
|
Data
|
numbers or scores generated by research study, measurements collected in a research study
|
|
|
Constant
|
Does not differ in quality or quantity
|
|
|
Conceptualization
|
4 step process used to narrow down list of potential variables by identifying those that must be measured to get answers to our research question
|
|
|
Steps in Conceptualization Process
4 S's |
Select-most impt. Variable to study
State- what is mfeant by each variable Specify-how each variable is to be measured State the value- of categories or values t hat each variable can assume |
|
|
Operationalization
|
Specifying exactly how to measure the variable that we have conceptualized, survey/observation
|
|
|
Reliability
|
The degree of consistency of measurement
Addresses the question=To what degree does the measurement of a variable produce consistent results? |
|
|
Validity
|
The desire to whixch a measurement instrument accurately measures what it claims to measure* using measurement instruments
|
|
|
Research Hypothesis
|
A statement of a relationship btwn. or among variables-often stated in the future tense because it predicts what will be found-expresses what we believe to be true
Eg. Smoking Pot (IV) compromises the brain Ability to function properly (DV) |
|
|
Independent Variable (IV)
|
The variable that is predicted to do the influencing
|
|
|
Dependent Variable (DV)
|
Believed to be influenced by the (IV)
|
|
|
Predictor Variable (PV)
|
Used for prediction
|
|
|
Criterion Variable
|
(Outcome Variable) Variables whose values we hope to predict
|
|
|
Discrete Variable
|
Can take on only an finite number of variables whose values
|
|
|
Dichotomous Variable
|
Has 2 value categories
Eg. Gender, T/F, M/F |
|
|
Binary Variables
|
Assigns numerical value to categories of 1 or 0 to indicate presence
Used when coding |
|
|
Dummy Variable
|
Created by converting a qualitative variable into a binary variable breaks it into sub-sets
Eg. Gender=male & female |
|
|
Descriptive Analysis
|
Nominal variables
Reduce large amounts of data to a simpler form, easier to understand |
|
|
Inferential Analysis
|
Ratio Variable
Used when *sample* is drawn from population & not the *total* population |
|
|
Frequency
|
Number of observations falling into a cell or value category of a specific variable
|
|
|
Frequency Distribution
|
Table or graph, shows the # of times (frequency) w/which different variables occur in a group of observations
|
|
|
Absolute Frequency
|
Number of times each value occurred *this total should always equal the last cumulative frequency number
|
|
|
Cumulative Frequency
|
Adds the absolute value together
|
|
|
Absolute Percent
|
Uses the absolute frequency and divides it by 10 and then multiplies it by 100, turn it into %, should equal 100
|
|
|
Cumulative Percent
|
Adds the absolute %, always want it to be 100%
|
|
|
Valid Percent incudes what?
|
The missing values, *Percent* does not.
|
|
|
Grouped Frequencies
What is meaningful group? |
Reduces the # of values to a smaller #, easier to understand, while not loosing meas
|
|
|
Descriptive Analysis
|
Nominal variables
Reduce large amounts of data to a simpler form, easier to understand |
|
|
Inferential Analysis
|
Ratio Variable
Used when *sample* is drawn from population & not the *total* population |
|
|
Frequency
|
Number of observations falling into a cell or value category of a specific variable
|
|
|
Frequency Distribution
|
Table or graph, shows the # of times (frequency) w/which different variables occur in a group of observations
|
|
|
Absolute Frequency
|
Number of times each value occurred *this total should always equal the last cumulative frequency number
|
|
|
Cumulative Frequency
|
Adds the absolute value together
|
|
|
Absolute Percent
|
Uses the absolute frequency and divides it by 10 and then multiplies it by 100, turn it into %, should equal 100
|
|
|
Cumulative Percent
|
Adds the absolute %, always want it to be 100%
|
|
|
Valid Percent incudes what?
|
The missing values, *Percent* does not.
|
|
|
Grouped Frequencies
What is meaningful group? |
Reduces the # of values to a smaller #, easier to understand, while not loosing measurement precision
|
|
|
Variability
|
Indicates the degree of variation among and value categories
|
|
|
5 different measures of variability
RIMVS |
Range
Interquartile range Mean deviation Variance Standard Deviation *Uses only Interval & Ratio to measure variability *Communicated best in a frequency distribution or graph, such as a bar chart |
|
|
Range
|
The distance the encompasses all values within the data set
Expressed as a formula: Range=maximum value-minimum value +1 |
|
|
Maximum Value
|
The value of the case w/ the largest value of the variable
Max age=35 MIn age=30 The range would be 6 (35-30+1=6) There are potentially 6 different ages(or values) that are included w/in the range: 35,34,33,32,31 & 30 |
|
|
Minimum Value
|
The value of the case w/the smallest value of the variableMax age=35
MIn age=30 The range would be 6 (35-30+1=6) There are potentially 6 different ages(or values) that are included w/in the range: 35,34,33,32,31 & 30 |
|
|
What is an Outlier
|
Data points that are far removed from and numerically distant from the rest of the points
|
