Term
| What is the principle of dilution in correlation? |
|
Definition
• A variable waters down the magnitude of effect of the other variable with each cross product
• Example: an imperfect 2:1 correspondence between X and Y (as opposed to perfect 1-1) |
|
|
Term
| What is the Correlation Coefficient? With what concept should correlation not be confused? |
|
Definition
• Correlation Coefficient: the degree of variation in one variable estimated from knowledge about another variable's variation
§ Can range from -1.0 to 1.0 (covariance between two variables converted to Z scores)
• Correlation (r) should not be confused with causation
§ Scores on x allow prediction of Y; as r's magnitude increases, so does our prediction precision |
|
|
Term
| hat is the Pearson product moment correlation? What meaning do the values -1.0 to 1.0 have? |
|
Definition
• Most commonly used index of correlation, appropriately used when both variables are continuous
* -1.0 to 1.0 indicate magnitude and direction of relationship |
|
|
Term
| What is the principle of least squares? How does it relate to the regression line? |
|
Definition
| • Regression refers specifically to prediction, and ximple bivariate regression uses one predictor variable (x) and one dependent (Y/criterion) variable |
|
|
Term
|
Definition
• Regression equations generate a predicted value of Y for each value of X and very rarely will these be exact (some error in prediction)
• Residual: the amount that the observed Y differs from the predicted value of Y' (Y-Y'; amount of observed Y 'left over' (residue)) |
|
|
Term
| What is the standard error of estimate? What is its relationship to the residuals? |
|
Definition
• Standard Error of the Estimate: the standard deviation of residules
§ An index of the accuracy of prediction
§ Small residuals are close to regression line and generate a small standard error of the estimate
§ Large residuals are far from regression line and generate a large standard error of the estimate |
|
|
Term
|
Definition
• Validity Shrinkage: the decrease in regression coefficients and coefficients of determination that result when exported outside their original group
§ Regression analysis overestimates the relationships strength between variables because it takes chance relationships into account along with the real ones, so, when it is exported to a new group of subjects, the chance factors drop out and reveal the true (weaker) relationsihps
• Shows that replicated studies tend to more clearly identify the strength of the actual relationships between variables than 'first time' studies |
|
|
Term
| What is restricted range? To what does it lead? |
|
Definition
• Restriction of Range (or Truncation): when the range of variables (X, Y, or both) becomes restricted
• Leads to reduced variance and may significantly alter the strength of the correlation coefficient
§ If variables X and Y each have ceiling effects, they may create a scatterplot that is compacted at the high end reducing the magnitude of correlation between X and Y |
|
|
Term
|
Definition
• Sudies interrelationships among a set of variables without reference to an external criterion
• A data reduction technique
§ Identifies the number of underlying constructs (factors) that the variables in a correlation matrix are collectively measuring
• Operates according to following…
§ When two variable in a correlation matrix correlate highly, they measure the same thing--they load onto the same factor
§ When they do not correlate, they load on different factors
§ One factor is extracted/named for each 'cluster' of highly intercorrelated variables in a correlation matrix
• Factors are defined by item loadings
§ Naming factors can be an inductive, rather than deductive process
§ One first evaluates the composition and relative sizes of the variables that "load" onto each factor before deciding what it measures; for this reason, this form of factor analysis is regarded as an exploratory statistical tool |
|
|
Term
| What is the co-efficient of determination? What is the purpose of the co-efficient of determination? |
|
Definition
• Index of…
§ The amount of shared variance between the variables X & Y
§ Degree to which the variation scores on X 'explains' the variation in scores on Y
§ Not an index of how much X causes Y
• How much scores of X explain the variation in scores on Y
• r2 = 0 (0%), .5 (50%), 1.0 (100%)
• Remember that the variance of one variable is "explained" by another variable to the extent that the variables covary or correlate |
|
|
Term
| What is the difference between simple linear regression and multiple regression? |
|
Definition
• Simple linear regression predicts one continuous dependent variable by a single predictor variable
• Multiple regression predicts one continuous dependent variable using a linear combination of two or more predictor variables (each one assigned weight based on the unique information that a single predictor contributes to the prediction of Y |
|
|
Term
|
Definition
• Test: a measurement device or technique used to quantify behavior and/or understand and predict behavior
• Item: specific questions or problems that make up a test; specific stimulus to which a person responds overtly; can be scored/evaluated |
|
|
Term
| Be able to define, recognize, and differentiate between states and traits |
|
Definition
• Trait: relatively enduring disposition
• State: mood; affect; temporary |
|
|
Term
| Define achievement, aptitude, and intelligence testing. |
|
Definition
• Achievement: previous learning
• Aptitude: potential for learning and acquiring a skill
• Intelligence: General potential to solve problems, adapt to changing circumstances, think abstractly, and profit from experience |
|
|
Term
| If a test is reliable its results are what? |
|
Definition
| • Reliability refers to the precision, dependability, consistency, or repeatability of test results; thus, a test is reliable if its results reflect these guidelines |
|
|
Term
|
Definition
| • Test batteries: two or more tests used in conjunction |
|
|
Term
| Define standardization? Why is it important to obtain a standardization sample? |
|
Definition
• Standardization: In statistics, a standard score indicates how many standard deviations an observation or datum is above or below the mean. It is a dimensionless quantity derived by subtracting the population mean from an individual raw score and then dividing the difference by the population standard deviation. This conversion process is called standardizing or normalizing; however, "normalizing" can refer to many types of ratios
○ A standard, normalized unit that has no dimension
It is important because it allows comparison of data from separate occasions |
|
|
Term
| Define representative sample and stratified sample. Know when and why representative and stratified samples are collected. |
|
Definition
• Representative Sample: one that comprises individuals similar to those for whom the test is to be used
• Stratified Sample: the population is divided into strata and a random sample is taken from each stratum according to the proportions of occurrence in a population |
|
|
Term
|
Definition
• Operational Definition: defining a way to measure a hypothetical construct; always one degree removed from the actual hypothetical construct
○ Measuring (with imprecision and error) the measurable phenomena to which the construct gives rise, but never the actual construct |
|
|
Term
|
Definition
| • Measurable Phenomena: quantifiable phenomena that can be directly measured |
|
|
Term
|
Definition
• Hypothetical Construct: processes that are not directly measurable, but which are inferred to have real existence and to give rise to measurable phenomena.
○ Because constructs are measured by operational definitions, they are (by extension) defined by those operations |
|
|
Term
| What is the difference between structured and projective personality tests? |
|
Definition
• Structured Personality Test: provides a statement, usually of the 'self-report' variety, and require the subject to choose between two or more alternative responses
• Projective Personality Test: unstructured; either the stimulus or the required response (or both) are ambiguous |
|
|
Term
| Define psychological testing and psychological assessment. How are they different? |
|
Definition
• Psychological Testing: the process of obtaining information
○ All the possible uses, applications, and underlying concepts of psychological and educational tests
• Psychological Assessment: the gathering and integration of psychology-related data for the purpose of making a psychological evaluation |
|
|
Term
| What is psychometry? What are the two major properties of psychometry? |
|
Definition
• Psychometry: the design of psychological tests to measure intelligence, aptitude and personality and the analysis and interpretation of the results
○ The branch of psychology dealing with properties of psychological tests
• The two major properties are Reliability and Validity |
|
|
Term
| What are norm- and criterion referenced tests? How is each unique? |
|
Definition
• Norm Referenced: Compare a test takers performance with others (usually people in same kind of group)
§ Class standing
§ Ranking
§ Percentile rank
• Criterion Referenced:
§ Measures performance against an established criterion
§ E.g., 70% score is a passing criterion; see whether or not the material was learned
§ Predicts performance outcome outside of the test (criterion referenced predictions) |
|
|
Term
| What types of questions are answered by psychologists through assessment? |
|
Definition
• Diagnosis and treatment planning;
• monitor treatment progress;
• help clients make more effective life choices/changes
• Program evaluation
• Helping third parties make informed decisions |
|
|
Term
| . In what settings do psychologists assess and what is their primary responsibility in each? |
|
Definition
• In and out patient hospitals and clinics
§ Diagnosis of pathology
• Schools
§ Diagnosis of learning disabilities, mental retardation, ADHD, giftedness, behavioral problems
• Forensic (legal) settings
§ Insanity defenses
§ Competency to stand trial
§ Psychopathology and need for treatment
§ Justification for lawsuits
§ Court-ordered evaluations (child custody/abuse)
• Employment settings
§ Applicant/employee fit for the job
• Career counseling
§ Primary assessment questions: interests, abilities, aptitudes |
|
|
Term
|
Definition
| § Measuring Scales: a set of numbers whose properties model empirical properties of the objects to which the numbers are assigned; any progressive series of values or magnitudes according to which a phenomenon can be quantified |
|
|
Term
|
Definition
| property of 'moreness'; when an instance of the attribute represents more, less, or equal amounts of the quantity |
|
|
Term
|
Definition
| : the difference between two points on a scale has the same meaning as the difference between two other points that differ by the same number of units § Relationship can be depicted by a straight line or linear equation; an increase in equal units on a scale reflects the equal incresaes in the meaningful correlates of units |
|
|
Term
|
Definition
| when nothing of the measured property exists (extremely difficult and rare for psychological qualities) |
|
|
Term
|
Definition
| • Discrete vs. Continuous: discrete is categorical and continuous is values may theoretically bet divided into progressively smaller units |
|
|
Term
| Know the four scales of measurement and be able to differentiate between these scales |
|
Definition
• Nominal: categories; values symbolize category membership and
§ can be classified, counted, and proportioned;
§ cannot be ranked, added/subtracted, or divided
• Ordinal: Rank; assignment of ranks according to the degree to which the measured attribute is present/absent
§ Can be classified, counted, proportioned, rank-ordered
§ Cannot be treated as if have equal intervals, so no arithmetic or division to form averages/ratios
• Interval: adjacent values on the scale represents equal intervals in magnitude of the attribute being measured
§ Can be classified, counted, proportioned, rank-ordered, added, subtracted, divided to form averages (mean)
§ Cannot be divided to form ratios (IQ 140 is not twice as intelligent as IQ 70)
• Ratio: measured on a scale with a true 0 point; allows all mathematical operations |
|
|
Term
| Define frequency distribution and histogram? What kind of data are shown in each? |
|
Definition
• Frequency distribution: displays scores on a variable or measure to reflect how frequently each value was obtained ; a tabulation of the values that one or more variables take in a sample
§ Shows each and every count
• Histogram: a number of cases for each value, shows the proportion of cases fall into each of several categories; does not provide information about differences in a specific interval
§ X-axis: range of observed values subdivided into equal intervals (lowest to highest)
§ Y axis: column height represents how many times each value of the x-axis was observed |
|
|
Term
|
Definition
| percentage of test takers whose scores fall below a given raw score § Median is used to describe CT; assumes interval level |
|
|
Term
| Define central tendency. Know the three types of central tendency and how to calculate each. |
|
Definition
• Measures of central tendency (CT): indices of the central value or location of a frequency distribution with respect to the x axis
• Mean: mathematical average score in a distribution
• Median: Middle score in the distribution (50% above and 50% below)
§ Rank the scores from lowest to highest
§ If odd, middle score, even, average of middle scores |
|
|
Term
| Know the advantages and disadvantages of the different measures of central tendency and when to use them. |
|
Definition
• Mean:
§ Pros: takes each score into account, interval level stats
§ Cons: outliers have heavy effect and skewed distribution pulls the mean
§ Best for: interval/ratio-level data that is somewhat normally distributed
§ Bad for: skewed, with outliers, or nominal/ordinal level data
• Median
§ Pros: Not affected by outliers/skewness
§ Cons: only 50th% considered; ordinal level stat (limited)
§ Best for: outlier/skewed, ordinal data
§ Useful when 50% is interesting in own right
• Mode:
§ Pros: interesting stat, no matter level
§ Cons: nominal level; extreme scores may be highest rate
§ Best for nominal data |
|
|
Term
| Define variance and standard deviation. |
|
Definition
• Variability--something changes or is variable from observation to observation
§ Measures: how distribution scores are scattered or dispersed
§ Range: distance from lowest to the highest score (outliers can possibly stretch the range--distort)
§ Interquartile Range: discards the distributions upper and lower 25%; IQR = Q3 - Q1 (could discard too much data)
§ Average deviation
• Variance-- the sum of squared deviations over the sample size
§ Sum of averaged squared deviations around the mean
§ Squared because it gives a value (not zero) and permits many least squares statistics
• Standard deviation: the square root of the sum of deviations squared over sample size (square root of variance)
§ Not an average deviation but gives approximation of how much a typical score is above or below the average
§ For samples, s, not sigma |
|
|
Term
| What is a z score? How is it calculated? |
|
Definition
• Standard scores: Raw scores that have been transformed from one scale to another with an arbitrarily set mean and SD
• z-score: standard distribution with an mean of 0 and SD of 1
§ Takes how scores are distributed around the mean to allow comparison of scores within 1 test or between 2+ tests (percentiles (1 = 84%, 2 = 98%) or SD units Z score 1 = 1 SD)
§ Answers how a score is distributed around the mean
• Difference between the score and the mean, divided by the SD
§
§ Converts to SD units by dividing by the SD |
|
|
Term
| How are T scores different from Z scores? |
|
Definition
• t-score: mean of 50 and an SD of 10
§ To convert from Z-scores: T = 10Z +50
• Similar:
§ have same shape as parent raw score distribution
§ Linearly transformed scores
§ Impart two pieces of information: location in relation to the mean & distance from mean
• Different:
§ Mean = 50 and SD = 10
§ Range from 5 SD above (T = 100) to 5 SD below the mean (T = 0)
§ Are all positive
§ Values +2 SDs are 'clinically significant' |
|
|
Term
|
Definition
| refers to divisions of the percentile scale into groups of equal fourths; points that divde the frequency distribution into equal fourths |
|
|
Term
| Define norm, norming, and standardization. For what is each used? |
|
Definition
• Norm: performances by defined groups on particular tests; a distribution of values constituting the typical performance of a given group
§ Based on score distribution obtained by a defined sample of individuals
§ Information about performance relative to a standardized sample
§ Data from a normative sample is used as a reference when evaluating and interpreting individual test scores
§ Norms describe the performance on a given test of a normative sample and provide context for interpeting test takers' scores
• Norming: the process of creating norms |
|
|
Term
| Define and differentiate between norm-referenced and criterion-referenced tests |
|
Definition
• Norm-Referenced Testing: compares each person to a norm; derives meaning by comparing scores to the normative sample
• Criterion Referenced Testing: references scores to some external standard (criterion); established based on values and standards of the test consumers
§ Sometimes 'domain-referenced' or 'content-referenced' tests due to their focus on a specified content are or domain
§ Assess achievment or mastery (curriculum-based assessment)
§ Disadvantage: performance relative to others is lost due to a lack of normative information (ex. Pass/fail or 98th vs 50th %ile) |
|
|
Term
| What is a scatterplot (scatter diagram)? How does it work? |
|
Definition
• Scatterplot: a graphical tool to identify linear and nonlinear relationships
§ Useful for: revealing curvilinear relationships, detecting outliers, detecting restricted range problems (ceiling effects, floor effects, central tendency rater errors)
• Bivariate: a scatterplot where:
§ One variable (x) is plotted on the x axis
§ The other variable (y) is plotted on the y axis
§ Visual data can be inspected to determine the degree of linear relationship |
|
|
