Download Psychology Assessment Methods: Interviews, Tests, Observations and more Exams Physics in PDF only on Docsity!
PSY 6645 Midterm latest update!!!
Assessment - •A comprehensive process that integrates information from multiple data collection methods. Three broad categories for collecting assessment information. - 1. Interviews
- Tests
- Observations Purposes of assessment? - 1.Screening 2.identification and diagnosis; 3.intervention (treatement) planning
- progress and outcome evaluation. Screening - Involving a single procedure or instrument, used to determine whether and individual has a high risk of having a specific problem and needs more in-depth assessment at that time. diagnosis - detailed analysis of an individual's strength and weaknesses with the general goal of arriving at a classification decision. intervention (treatment) planning - involves deciding on a course of action that facilitates client change and helps improve the client's outcome. Tests - These are instruments designed to measure specific individual attributes. These instruments can reveal important diagnostic information not otherwise identified through other assessment methods Observation - •Mechanism for observing and recording behavior in a particular environment or setting. Assessment process: - 1.Identify the/a problem 2.select and implement assessment methods 3.evaluate obtained assessment information
- report assessment results and make recommendations Qualifications to use test measures: - Responsibilities of Users of Standardized Tests (RUST) Responsibilities of Users of Standardized Tests (RUST) (4 Qualifications) - 1. Purposes of testing
- Characteristics of tests
- Settings and conditions of test use
- Roles of test selectors, administrators, scorers and interpreters. computer administration of assessment instrument - Administering tests, question- naires, and interviews via the computer is one of the most common computer assess- ment applications. This response format has many advantages over traditional paper- and-pencil methods, such as increased delivery, the potential time savings, and the ability for items to be adapted or tailored based on the test taker's response to a previous item. Automated Test scoring - Computer-based assessment provides automated scoring of responses, thereby giving test takers almost immediate feedback and their overall score. Computer scoring reduces the possibility that respondents would make errors while filling out handwritten answer sheets, and it eliminates the errors that clinicians and technicians would make while hand-scoring items. computer-generated reports and narratives - Computer-based assessment instru-ments often provide computer-generated reports or narratives. These reports are auto-mated interpretations that are generated based on user input and resulting test scores (Butcher, 2013). The reports may contain very complex and detailed statements or summary statements. Computer adaptive test - Computer-adaptive tests are tests that are specifically tai-lored to an individual's ability level. The computer quickly locates the examinee's ability level and then tailors the questions to that level. The first question is usually selected close to the passing level. If the test taker answers the question correctly, a more difficult item is presented next. Using a computer-adaptive test, test takers have a more personalized assessment experience in a controlled environment. Computer-adaptive tests also provide sensitivity to the needs of users with disabilities, helping ensure equality and fairness in testing. computer simulation - Computer simulation is the technique of representing real-world experiences through a computer program. Interactive software programs allow individuals to explore new situations, make decisions, acquire knowledge based on their input, and apply this knowledge to control the ever-changing simulation state. Simulations have been in use for many years to assess performance in different envi- ronments. In the military, simulation has long been used for assessing the readiness of individuals to perform military operations, and devices used for computer simula-tions range from plastic mock-ups to laptop computers to full-motion aircraft simula-tors. In education, simulations can be used to investigate problem-solving skills, allowing students to explore a range of options in a particular problem scenario. Scenario-based testing is also used for some counseling-related exams such as the Clinical Mental Health Counseling Exam (CMHCE). unstructured interview - Very flexible Informal (nonstandardized) Interviewer may follow a general format Widely used
audio administration - The test is presented using audio recordings. American Sign Language - The examiner gives direc-tions and presents the test items using sign lan-guag nonverbal - The examiner avoids oral or written directions and relies on universal hand and body gestures to explain tasks to the exam-inee. Observation methods - Formal and informal observation direct and indirect observation; naturalistic and formal observation - is a highly structured process in which the observer decides ahead of time who will be observed, what behavior will be observed, when and where the observation will take place, and how the behavior will be recorded. direct observation - observations of behavior are not filtered through the perceptions of some informant—the individual's behaviors are observed directly indirect observation - rely on reported observations of behaviors by others who have direct contact with the individual. naturalistic observation - Observations that take place in naturally occurring settings, such as an individual's work or school, contrived observation - Observations may also occur in a laboratory or other unobtrusive - when there is no interaction between the observer and those being observed, and when the individual's behavior is not affected by the observation itself. O particpant observation - entails the observer both watching and interacting with the individual as part of the observational situation. Methods of recording - 1. event recording
- duration recording
- time sampling (interval recording) event recording - is the simplest of the observation data collection methods. It requires an observer to observe, count, and record the number of times a behavior has occurred. duration recording - is used when it is more important to know how long a behavior occurs rather than the frequency of the behavior. time sampling (inteval recording) - in both event and duration recording techniques, all occurrences of the behaviors are documented during the observational period. H
types of rating scales - 1. likert
- graphic
- semantic differential graphic scale - is similar to the Likert scale, except it presents respondents with a graphic five-or seven-point continuum rang-ing from "never" to "always" or from "strongly disagree" to "strongly agree." Semantic differential scale - consist of bipolar adjectives separated by a seven-point scale on which respondents select one point to indicate their response. likert - named after the psy-chologist who developed it) consists of a series of written statements to which respondents indicate how much they agree or disagree (i.e., Strongly Disagree, Disagree, Neither Agree nor Disagree, Agree, or Strongly Agree). T Leniency or generosity - Ratings tend to always be at the top of the range. Severity - Ratings tend to always be at the lower end of the scale. Central tendency - Ratings tend to fall consistently in the middle or average range. Response acquiescence - Rater tends to agree with each item. Response deviance - Rater tends to respond in a deviant, unfavorable, uncommon, or unusual way. Social desirability - Test responses are interpreted to provide the most favorable view of the examinee. Halo effect - Ratings tend to be influenced by good impressions of the examinee. High ratings on some traits are generalized to others. Negative halo - Ratings tend to be influenced by bad impressions of the examinee. Logical error - Ratings are influenced by the person's most recent performance rather than their usual level of performance. Contrast error - Ratings on the current candidate are influenced by the previous candidate. Proximity error - Ratings made on items that are printed close together tend to be rated the same. Most recent performance error - Ratings are influenced by the person's most recent performance rather than their usual level of performance.
c level - instruments require users to have B-level qualifications plus a doc-torate degree in psychology or a related discipline (that provides appropriate train-ing in the administration and interpretation of assessment instruments) or licensure/certification, or be under direct supervision of a qualified professional in psychology or a related field. Examples of C-level instruments include intelli-gence tests, personality tests, and projective measures (e.g., the Wechsler Intelli-gence Scale for Children [WISC-V], the Minnesota Multiphasic Personality Inventory [MMPI-II], and the Rorschach Inkblot Test). Test Administration - Important to inform test taker about test instrument(s), battery, etc. In orientation include purpose of assessment instruments, criteria used by examiner in selecting specific instruments, conditions in which the instrument will be taken, range of skills or areas measured, administrative procedures and concerns (cost, group vs individual, time involved), overview of test questions, types of scores, and process of reporting assessment results to the test taker scales of measurement - nominal, ordinal, interval, ratio nominal - Names, lists of words Ordinal - Rank ordered, Likert scales interval - Temperature ratio - Age, height, weight describing sets of scores - • Frequency distributions
- Measures of central tendency
- Measures of variability •Measures of relationship frequency distribution - •way of ordering scores and data; can further group scores within distribution by using class interval •Graphs, histograms, frequency polygon can be used to organize and describe scores •Frequency polygon can be symmetrical distribution or asymmetrical (skewed) distribution. Kurtosis - •a statistic that describes the peakedness or flatness of a distribution. Uses the normal distribution as the framework for such description. mesures of central tendency - •what is the typical or average score within a score distribution mean, median, mode measures of variability - degree that the scores are spread out (dispersed) from each other within a distribution of scores
mean - the most frequently used measure of central tendency, is the arithmetic aver- age score in a distribution. Median - the middle score, or the score that divides a distribution in half; 50% of the scores will fall below the median and 50% will fall above. mode - is the score or numerical value that appears most frequently in a set of scores. types of correlation formulas - •Pearson Product Moment correlation •Spearman Rho correlation •Phi Coefficient correlation Scatterplots a graphical display showing the spread of scores for each variable. pearson Product Moment correlation - repre-sented by the lowercase r, is used for measuring a linear relationship between two con-tinuous variables, such as weight and intelligence in children. Spearman Rho correlation - is a variant of Pearson's r that is used for finding the association between two ordinal (rank-ordered) variables. Phi Coefficient correlation - (f) is used to assess the degree of association between two nominal variables. Scatterplots - Calculating the correlation coefficient should be undertaken in conjunction with a scatterplot of the two variables. Regression - •uses correlation to predict an outcome or relationship. Relationships among variables (can use more than two variables) are analyzed to determine how a variable may predict another variable (or variables) Types of regression - •include simple linear regression (the value of a dependent variable is predicted by another variable-independent variable) and multiple regression (a dependent variable is predicted from two or more independent variables). factor analysis - the relationships among a large number of variables are explained by common underlying dimensions or factors. Is a means of reducing data to common denominators or factors. Commonly used in developing construct validity for test measures. charcteristics of test scores - 1. criterion-referenced
- norm referenced criterion-referenced - What the test taker knows or does not know with respect to a specified content domain.
Qualitative descriptions - narrative descriptions are used to describe test performance. Usually a narrative description (qualitative description) is used to describe a particular score range (quantitative performance). example IQ 130 and above 120-129 110-119 90-109 80-89 70- 69 and below Classification Very Superior Superior High Average Average Low Average Borderline Extremely Low tables - are often provided to aid in score interpretation. profiles - also help in interpreting results by providing a visual representation of scores. raw - Gives precise number of points scored on a test. Easy to compute Stanine - One-digit scores Can be averaged Used for simplicity and utility Reliability - •The precision or dependability of test scores factors that cause measurement error - time sampling length of test content sampling test taker variables inter-rater differences test administration factors test item factors Methods for Estimating Reliability - test-retest alternate forms (parallel forms) internal consistency (split-half, Kuder-Richardson formula, Coefficient Alpha) inter-rater test-retest - Same test given twice with time interval between testings Simultaneous Administration (alternate form) - Equivalent tests given at the same time Delayed Administration (alternate form) - Equivalent tests given with a time interval between testings
split-half (internal consistency) - One test is divided into two compa-rable halves, and both halves are given during one testing session KR Formulas and Coefficient Alpha (internal consistency) - One test given at one time (items compared to other items or to the whole test) Interrater - One test given, and two individuals independently score the test standard error of measurement - fluctuations in individual's test scores if examinee retook test on multiple occasions.
