Reviewer Psych Assessment P1-P3 PDF

REVIEWER PSYCH ASSESSMENT P1-P3 TESTING AND ASSESSMENT 2. STANDARDIZED Psychological test/testing - Uniformity exists in administration, - A test is a measuring device, procedure, scoring, and interpr...

REVIEWER PSYCH ASSESSMENT P1-P3 TESTING AND ASSESSMENT 2. STANDARDIZED Psychological test/testing - Uniformity exists in administration, - A test is a measuring device, procedure, scoring, and interpretation of or technique used to quantify behavior psychological test data. - Test or procedure designed to measure psychological variable (personality, aptitude, intelligence, attitudes, and 3. RELIABLE values) - There is consistency in test results. - We only measure a sample behavior ng isang tao 4. VALID Duration: Shorter lasting from few minutes to - The test measures what it aims to few hours measure. Sources of data: one person (test taker only) Qualification use: knowledge of tests and TOOLS OF PSYCHOLOGICAL ASSESSMENT testing procedures Cost: expensive 1. Interviews - method of gathering information through direct communication involving reciprocal Psychological Assessment Duration: Longer lasting a few hours,to a few 2. Portfolio days or more. - flies of work products; samples of one's Sources of data: collateral sources (relatives, abilities and accomplishments are used teachers, etc..) as assessment tools. Qualification use: knowledge of testing and other assessment methods as well as of 3. Case History Data speciality area assessed (psychiatric disorder, - records transcripts, and other accounts job requirements, etc..) in written, pictorial or other form that Cost: very expensive, requires highly qualified preserve the archival information, official professionals and informal accounts and other data and items relevant to the assessee CHARACTERISTICS OF PSYCHOLOGICAL TESTS 4. Behavioral Observations - monitoring the actions of others or 1. OBJECTIVE oneself by visual or electronic means - Psychological test results are free from while recording qualitative/quantitative the subjective perception of the test information regarding those actions administration. - the tendency to base judgments and TYPES OF PSYCHOLOGICAL TESTS interpretations on external data rather than on subjective factors, such as I. According to the number of test takers: personal feelings, beliefs, and experiences; Individual Tests  Group Tests administered, scored and interpreted without specialized training, by following the instructions II. According to the variable being in their manually measured: LEVEL B (Intermediate) Abilly Tools that are for some specialized training in Achievement test construction and use and in the area in Aptitude which the Instruments will be applied such as Intelligence aptitude tests and personality Inventories Personality applicable to normal population. Objective or Structured Projective or Unstructured LEVEL C (Highest) Instruments that require extensive familiarity PARTIES IN PSYCHOLOGICAL with testing and assessment principles, as well ASSESSMENT as with psychological fields to which the instrument pertain such as individual Intelligence Test Authors/Developers tests and projective technique - create test or other methods of assessment RELIABILITY Test Publisher - they publish, market, and sell tests, - Consistency of measure thus controlling their distribution Parallel forms reliability estimate Test Reviews - Compare two equivalent forms. These - they prepare evaluative critiques of tests two forms use different items but with based on their practical and technical the same level of difficulty merits Alternate forms Test Users - Are simply the different versions - professionals such as clinicians, of the test that have been psychologists, etc. use these tests for constructed so as to be parallel assessment Low Reliability Test Sponsors - Increase number of items - Institutional boards or government - Use factor analysis agencies who contract test developers - Use the correction of attenuation or publishers for various testing formula (used to determine the exact services. correlation between two variables How Reliable is Reliable? THREE TIER SYSTEMS OF - Test-retest and alternate forms= the PSYCHOLOGICAL TEST more the better internal consistency LEVEL A (Lowest) Basic research.70 and.80 coefficient A limited range of Instruments such a (good enough) educational achievement tests that can be Clinical setting.95 reliability coefficient or above  - Measure one factor, such as Things to remember about test-retest one ability or one trait which reliability yields to high consistency > If the result of the first and second - Unidimensional administration has low a correlation, it might mean that: Heterogenous - The test has poor reliability - Measures more than one factor and result to low consistency Test - retest reliability estimate - Multidimensional - Correlating pairs of score from same people; two different administrations of Speed test vs Power test the same test - Useful only to traits or characteristics Speed test that do not change over time; IQ for - The goal is to evaluate speed instance; coefficient of stability only, or a combination of speed and correctness Sources of error variance - Questions are relatively easy in Test constructions nature - Item sampling and content sampling is one source of error in test constructions Power test - The goal is to evaluate Test administration correctness in the sense - The manner of test administration may knowledge, skills, and abilities influence the test takers of test-takers - Questions are relatively difficult VARIABILITY in nature - is an indication of how scores in a distribution are scattered or dispersed. Dynamic vs. Static characteristics Range Dynamic - known as the simplest measure of - Presumed to be an ever variability changing characteristic that changes through time or Standard Deviation situation as a result of - quantifies the amount of dispersion experiences. For instance trait, of data values around the mean. state, or ability. - Internal consistency is of best use. Variance - capable of action or change - measures the average squared distance that scores from their Static mean. - Characteristics that would not vary across time or situation THE NATURE OF THE TEST such as intelligence. - Fixed Homogenous vs Heterogenous Homogenous  Graduate Record Exam (GRE) or Scholastic Aptitude Test (SAT) NORMS - Used for admission to graduate school - Process of deriving norms and college - Mean500; SD = 100 Normative sample - A normative sample is a group of Stanine individuals used as a standard to - Takes whole numbers from 1 109 compare and evaluate the performance without decimals. or behavior of other individuals. - Mean 5: SD=2 - Used by US Air Force Assessment. Race Norming - Norming based on the race or ethnic background STANDARD SCORES User/program norms - Norms provided by the test T-SCORE manual - Can be called a fifty plus or minus ten scale. Types of norms referenced testing - Mean50: SD=10 - Devised by McCall in honor of his Developmental Norms professor E.L Thorndike (1). - Normed developed on the basis of any trait,ability, skill, or other characteristics that presumed to Z-SCORE develop, deteriorate, or - statistical measurement of a score's otherwise be affected by relationship to the mean in a group of chronological age,school grade, scores. or stage of life - MeanD: SD-1 Within Group Norms - Zero plus or minus one scale - Individuals performance is - When determined, it can be used to evaluated in terms of the translate one scale to another, "golden performance of the most nearly scores” comparable standardization group. CORRELATIONAL STATISTICS - Artificial tools for testing the OTHER STANDARD SCORE relationship or association between variables Deviation IQ - Used for interpreting IQ scores. - Mean 100; SD = 15 Covariance: how much true scores vary together Sten Correlation Coefficient: - Standard ten mathematical index that - Mean 5.5; SD = 2 describe the direction and magnitude  - Always ranges from - For example, score in the test -1.00 +1,00 only (continuous or interval) and correctness in an item within the test (true Phi coefficient dichotomous). - Correlates two dichotomous data; at least one should be a true dichotomy. Spearman Rho Correlation - For example, the gender population who - A method of correlation for finding the passed or failed the 2018 Physician association between two sets of ranks Licensure Exam. thus, two variables must be in ordinal scale. Tetrachoric coefficient - Also called rank-order correlation - Correlated two dichotomous data: both coefficient or rank-difference correlation. are artificial dichotomies. - Devised by Charles Spearman - For example, passing or failing a test and being highly anxious οr not THREE TYPES OF CORRELATIONS: True Dichotomy The positive correlation - There are only two possible categories - When one voliocle incomes the that are formed naturally. other variables also increase - For example: Gender (M/F) Negative correlation - When one variotike increases, Artificial Dichotomy the other variable decreole - can be created simply by comparing an No correlation interval scaled variable to a threshold - No movement or association was observed Biserial Correlation - Expresses the relationship between a Spearman Rho Correlation continuous variable and an artificial dichotomous variable. - A method of correlation for finding the - For example, the relationship between association between two sets of ranks passing or failing the bar exam (artificial thus, two variables must be in ordinal dichotomous variable) and general scale. weighted average (GPA) in law school - Also called rank-order correlation (continuous variable) coefficient or rank-difference correlation. - Devised by Charles Spearman Pearson Product Moment Correlation - Determines the degree of variation in one variable that can be estimated from Point Biserial Correlation knowledge about variation in other - Correlates one continuous and one true variables. dichotomous data - Correlated two variables in interval or - For example, score in the test ratio scale format. (continuous or interval) and correctness - Devised by Karl Pearson in an item within the test (true dichotomous). Point Biserial Correlation - Correlates one continuous and one true dichotomous data ISSUES IN THE USE OF CORRELATION Residual: difference between the predicted and - Assumes that each person has a true the observed values. score that would be obtained if there were no errors in measurement. Standard Error of Estimate: standard deviation - The true score for an individual will not of the residual: measure of accuracy and change with repeated applications of the prediction. same test. Shrinkage: the amount of decrease observed when a regression equation is created for one THE NORMAL CURVE population and then applied to another. - A bell-shaped, smooth, mathematically defined curve that is Coefficient of Determination (12): tells the highest at its center, From the proportion of the total variation in scores on Y center, it tapers on both sides that we know as a function of information about approaching the X-axis X. It also suggests the percentage shared by asymptotically approaches, but two variables; the effect of one variable to the other. never touches, the axis Coefficient of Alienation: measures the NORMAL DISTRIBUTION: non-association between two variables. - majority of the test takers are bulked at the middle of the distribution; very few Restricted Range: significant relationships are test takers are at the extremes. difficult to find if the variability is restricted. Mean Median Mode Measurement - the act of assigning numbers of KURTOSIS symbols to characteristics of things - It is term that testing professionals (people, events, whatever) according to use to refer to the steepness of a rules. distribution in its center Scale - Describes how heavy or light the - a set of numbers (or other symbols) fails are. whose properties model empirical properties of the objects which the PLATYKURTIC - relatively flat numbers are assigned. gently curved MESOKURTIC - moderately curved Continuous Scale used to measure somewhere in the middle variables that are continuous: weight, LEPTOKURTIC - relatively peaked height, temperature, are some of the examples. Discrete Scale used to measure CONCEPTUALIZATION OF ERROR variables that are discreet such as number of males in the class, coins in THE TEST SCORE THEORY jar, etc. Classical Test Score Theory: - Major assumption is that errors of Interval Scale measurement are random.  - There is a numerical distance between conceptually distinct and potentially each value. independent aspects of ratee's behavior. - A level of measurement at which it is General Standoutishness - people possible to average a set of tend to judge on the basis of an measurements and obtain meaningful outstanding characteristic. Example, a results well groomed applicant is Interpreted as intelligent. SKEWNESS: EXPECTANCIES: tendency for results to be - The nature and extent to which the influenced by what the fest administration symmetry is absent. expects to find. - An indication of how the measurements in a distribution is distributed Rosenthal Effect-the administrator's - Helps to identify the relationship expected results influence the result of between the mean, the median, and the the test. mode. Golem Effect - negative expectations from the test administrator decreases ITEM one's performance. Apostle which penan respond everty DECEPTION and this response can be scored or - the tendency of the observers to induce evaluated through the use of scale lies in order to gather a much more accurate sample of behavior. POSITIVELY SKEWED: RATING ERRORS - a judgment resulting from the When relatively few of the scores fall of intentional or unintentional misuse of a the high end of the distribution. Simply, rating scale. more lest takers got a low score Mean > Median > Mode Leniency Error or Generosity Error tendency of the rater to be too forgiving NEGATIVELY SKEWED: and insufficiently critical. Severity Error-rater is overly critical. When relatively few of the scores fall or Central Tendency Error the rater is the low end of the distribution. Simply, reluctant to give ratings at either positive more test takers got a high score. REACTIVITY - being evaluated Increases performance; ISSUES OF BEHAVIORAL ASSESSMENT also called Hawthome Effect. METHODS DRIFT RATING ERRORS - moving away from what one has learned going to Idiosyncratic definitions of Halo Effect -tendency to ascribe behavior, suggests that observers must positive attributes independently of the be retrained at a point in time. observed behavior, giving a particular ratee a higher rating than he/she Contrast Effect the tendency to rate the objectively deserves because of the same behavior differently when rater's failure to discriminate among  observations are repeated in the same Structured: provides a self-report statement to context which the person responds "True" or False, "Yes" or "No.". MEASURES OF PERSONALITY ACHIEVEMENT TESTS: PERSONALITY TESTS: Measure typical behavior-traits, temperaments, and dispositions. Measure previous leaming. Used to measure the effectiveness of an Structured: provides a self-report educational or training statement to which the person responds program Result of relatively structured "True" or False," "Yes" or "No.". input. Projective: provides an ambiguous test Could be factual or conceptual stimulus; response requirements are Example: Midterm Exam in unclear. Physiological Psychology Interest: measures likes and dislikes as well as one's personality orientation toward the world of work. GROUP TESTS: TESTS OF HUMAN ABILITY - Administered by a test administrator INTELLIGENCE TESTS: - Measure the potential to solve to a large number of people problems, adapt to changing preferably with a ratio of 1:25. circumstances, and profit from - Usually in a pencil and paper format. experience.. - Administered in educational and - Example: Culture-Fair Adaptability Test Industrial settings. - Information gathered is limited to the purpose of what the test intends to APTITUDE TESTS measure. Measure potential for acquiring new skills. Suggests how well a test taker might succeed in a certain area of vocation. Tend to focus on informal learning rather than formal learning. Also known as prognostic fest or readiness test and is used to make predictions and measures the readiness of the test faker to a certain program Example: Differential Aptitude Test PERSONALITY TESTS: Measure typical behavior-traits, temperaments, and dispositions.