Psych research methods exam 2 flashcards
What are common ways of establishing reliability?
- Alternate form reliability:
- Test-retest reliability
- Measures of internal consistency
- Interrater reliability
Alternate form reliability
When you develop 2 versions of a measure, give them both to the same group of people, then correlate them.
Test-retest reliability
When you have one version of exam, give it to the same group of people a
2 different intervals
Measures of internal consistency
Measures of much items correlate with one another
Split-half reliability
Cronbach’s alpha
Split-half reliability
data collected is split randomly in half and compared, to see if results taken from each part of the measure are similar
ex: Test split in half, class take each half and the results are correlated
Cronbach’s alpha
Average of all the ways you can split half an exam
Inter-rater reliability
Kappa
determine how much raters agree with each other
Reliability coefficient of ways to establish reliability
Reliability of coefficient of at least 0.7 for all the methods of reliability except for kappa
Kappa: 0.4-0.75 is good for reliability
Population
Collection of units to which we want to generalize a set of findings or a statistical model too
Representative sample
Smaller collection of units from a population used to determine truths about that population
Random selection
Randomly select people from a population
Random sample
Each individual of the population has an equal chance of being selected
In a random sample, the probability of people have to stay the same
Parameter
A value, usually a numerical value, that describes the entire population
The value is derived from measurements of the individuals in the population
Statistic
A value, usually a numerical value, that describes a sample
Derived from measurements of the individuals in the sample
Statistical notation
Scores are referred to as X and Y
Collecting data on age and number of Facebook friends, you label age as X and friends as Y
N refers to the number of scores in a population
n is the number of scores in a sample
Sigma (summatation)
Add a set of scores from the sample
Summatation
Done after operations
- in parenthesis
- Squaring
- multiplication
- division
Summation is done before
Addition
Subtraction
What is a frequency distribution table
Representation that tells the frequency of different scores in your sample
Function: Organizes all the data so that the complete distribution can be viewed all at once
What are examples of frequency distribution tables and what are they used for?
Polygon or histogram for continuous data
Bar graph for categorical data
How to differentiate between a bar graph and histogram
Bar graph has spaces between them and is only used for categorical data
while histograms are used for continuous data
Statistical notation
Scores referred to as x and y
N refers to number of scores in a population
n is the number of scores in a sample
Σ stands for summation stand for summation and is when the set of scores are added up
Summation
Represented by Σ
Done after PEMD
Done before AS
What are the measures of internal consistency
Split-half reliability
Coefficient/Cronbach’s alpha
Bar graph
graph that presents categorical data with rectangular bars with heights proportional to the values that they represent.
Histogram
chart that plots the distribution of a numeric variable’s values as a series of bars
What are the four characteristics to look for in a frequency distribution
- Shape
- Location
- Spread
- Sample size
Types of shapes in a frequency distribution
Modality
Symmetry
Skew (positive or negative)
Slope-Kurtosis (leptokurtic or platykurtic)
What is the location of a frequency distribution?
A measure of where the bulk of data sits on the number line and where the central tendency is
Central tendency
A statistical measure where a single score defines the center of a distribution
What is the purpose of the central tendency?
Describe the distribution by identifying its center
Find the single score that best represents the entire group
What are the measures of central tendency?
Mean
Median
Mode
What happens to the mean when you change the value of a score or add/remove a value?
The mean changes unless the score added or removed is equal to the mean
What happens to the mean when you add or subtract a constant from each score
Changes the mean by the same constant
What happens to the mean when you multiply or divide each score by a constant?
The mean would also be multiplied or divided by that constant
Line graph
Graphical representation of information that changes over a period of time
Used to show how something changes over time
George W. Bush’s Tax cuts
Tax deduction passed in 2001 that stated 92 million Americans would receive an average tax reduction of 1083
However the median tax cut was $100
Median
Midpoint of distribution (defined by number of scores)
Mean
Balance point of a distribution
What is common between the mean and median
Both measure central tendency
Mode
Most frequent score/value in the dataset
Only measure of central tendency that can be calculate categorical variables
Range
Difference between covered highest and lowest values
Considered unreliable measure of variability as it is based only 2 scores
What are the Quartiles
Three values that split sorted data into four equal parts
What are the three quartiles
Lower quartile = median of lower half of data
Second quartile = median
Upper quartile = median of upper half of data
Interquartile range (IQR)
The range of the middle half of the data
What is a box plot and what can it effectively identify
Visualization of median and IQR
Effectively identifies outliers
How is total error found
By adding up all the deviations
How is the Sum of Squared Errors found
- Add deviation
- Square each deviation
- Add all the squared deviations
What is the Definitional Formula of Sum of Squared Errors
SS = Σ(X–μ)^2
Steps
1. Find each deviation score
2. Square each deviation score
3. Sum up the squared deviations
What is the computational formula of sum of squared errors
SS = Σx^2 - (ΣX)^2/N
Steps
1. Square each score and sum them
2. Find sum of scores and square it, divide by N
3. Subtract the step #2 from the first step
Notation of Σ
Sum the scores
Notation of N
number of scores in population
Notation of n
number of scores in sample
Notation of X and Y
Each score is referred as X and Y
What is the variance
Statistical notation (s^2)
Calculated by dividing SS
Notation of SS
Sum of Squared Errors
Variance of sample formula
S^2 = SS/n-1
Variance of population
s^2 = SS/n
Notation of s
Standard deviation
Standard deviation of sample formula
s = square root (SS/n-1)
Standard deviation of population formula
s = square root (SS/N-1)
What is standard deviation
An average for how far data points are from the mean
A large derivative means more spread out
(determines representation of data)
What are things SS, S^2 and S all represent
- Fit of mean to data
- How well mean represents observed data
- Variability in the data
- Error
How would adding or subtracting a score by a constant change SD
The SD would stay the same
How would multiplying or dividing a score by a constant change SD
Causes the standard deviation to be multiplied by the same constant
Positive skew
The tail is more pronounced on the right side
Means skewedness is less than 0 since it is in the positive direction
Negative skew
The tail is more pronounced on the left side
Means skewedness is less than 0 since it is in the negative direction
Kurtosis
Bell curve shape in distribution tables that describes distribution of observed data around the mean
Leptokurtic
Low kurtosis
Characterized by broad peak and thin tails
Platykurtic
High kurtosis
Characterized by high peak with thick tails
Skewness
Measure of the asymmetry of a distribution
What is a correlation?
Way of measuring the extend to which two variables are related
What are the characteristics of a correlation
Direction (negative or positive)
Form (linear is most common)
Strength
What are the three measures of variability
Range
Interquartile range
Standard deviation
What are the functions of measures of variability
Describes whether the scores are widely scattered or closely scattered
Helps describe the location of individual scores
Gives indication how well a measure of central tendency represents the whole group
What is the correlation coefficient
Variance shared by both variables/(total variance of both variables)
= +1/-1
How much of total spread variance is shared between both variables
Correlation coefficient levels
0 = no relationship
+/- 0.1
What is the coefficient of determination?
Proportion that each of the variables play
r^2
Sum of products (SP)
Measures covariability between two variables
SP = Summation (X- Mx)(Y-My)
Problems with covariability
- Depends upon the units of measurement
- One solution
Partial correlation
Measures the relationship between two variables, controlling for the effect that a third variable has on them both.
Z score
Standardizing a score with respect to the other scores in the group.
Allows you to identify and describe location of every score in the distribution
Computing a z-score is equivalent to asking where the score (X value) is located in the distribution.
Z score calculation
Z = X - (mean of sample)/ SD
What does the sign and number of the z score tell you
The sign tells whether the score is above or below the mean
The number tells distance between score and mean in standard deviation units
What is a shape of a normal distribution?
- Symmetrical
- Highest frequency in the middle
- Frequencies taper off towards the extremes
Definition of probability
The likelihood that a specific outcome will actually occur
Probability formula
number of outcomes classified as A/total number of possible outcomes
ex. 4 men, 15 women
4/19