Slide 1Slide 2Slide 3Slide 4Slide 5Slide 6Slide 7Slide 8Slide 9Slide 10Slide 11Slide 12Slide 13Slide 14Psychometric Sampling3-way samplingSlide 17Slide 18Slide 19Slide 20Slide 21Slide 22Slide 23Slide 24Slide 25Slide 26Slide 27Slide 28Slide 29Slide 30Introduction to Psychological Measurement• Psychometrics & some vocabulary• Kinds of scale construction• Desirable properties of a psychometric instrument• Psychometric sampling• Kinds of items• Scaling modelsPsychometrics (Psychological measurement)The process of assigning values to represent the amounts and kinds of specified attributes, to describe (usually) persons.• We do not “measure people”• We measure specific attributes of a person Psychometrics is the “centerpiece” of empirical psychological research and practice.• All data result from some form of “measurement”• The better the measurement, the better the data, the better the conclusions of the psychological research or applicationA bit of vocabulary to get us started…Kinds of Items:• survey item – an individual item that will measure the target construct• scale item – one of a set of items that, when combined, will measure the target constructe.g., age vs. emotional maturity ScalesWe’ll use the terms “scale” to mean “a multi-item instrument designed to represent the amount or kind of a specific attribute for a specific individual”A bit of vocabulary to get us started…Population:The group of individuals to which we want to apply our scale. We sample participants to represent the population.Domain:The type of information we want to measure with our scale. We sample items to represent the domain.Construct value:The individual’s amount or kind of the attribute or characteristic we are trying to measure.Variable: The individual’s measured score or codeKinds of Scale ConstructionWe’ll use the terms “scale” to mean “a multi-item instrument designed to represent the amount or kind of a specific attribute for a specific individual”.So… all kinds of things are “scales” !!!The construction and validation of most scales follows a pretty similar overall logic and process. However, there are differences depending upon:• what content/construct we’re trying to measure• intellectual assessment, mental health evaluation, achievement testing, employment selection, research IVs/DVs, etc.• to what legal and organizational auspices we’re beholden• EEOC, APsychologicalA, APsyciarticA, EMAA, SIOP, etc.Kinds of Scale Construction, cont.Research scales – the emphasis in this class• sometimes you can’t find or can’t afford a measure of someconstruct or behavior you’d like to research (new scale)• sometimes you’ll want to capture a “different version” of a construct than do existing instruments (new scale)• sometimes you want to measure more things that you havetime to measure using their current version (short form)• sometimes you want an alternative version of a scale for pre-post or other repeated measures (alternate form)• sometimes you’ll want to “take apart” some construct or proxy variable, e.g., motivation (multiple subscales)Kinds of Scale Construction, cont.Selection Research – a “bonus” in this class• Whether for education, employment or recreation, there are usually more “applicants” than “opportunities”• “Fair” means of making these selections are increasingly required by organizations making the selections, by professional organizations and by the lawClassroom Tests – important for many of you and a useful counterpoint to research instruments• writing tests to maximize assessment accuracy, educational benefit, both??Another “kinds of Scales” – type of resulting informationClassificatory or Ordered Category scales• scales designed to divide people into those that have “enough” vs. “not enough” of a construct• e.g., pass – fail score Measurement or Equidiscriminating scales• scales designed to give a score that represents a person’s position along the contruct continuum• e.g. % correct score to represent “how much they know”Many scales are a designed to provide a “blend” of these….• to divide the construct continuum into multiple ordered categories• e.g. “A” “B” “C” “D” “F” grade scaleSometimes it is very hard to label the “type” of scale & measurement involved – take classroom grades…• start with # or % correct on tests, homework, etc – looks equidiscriminating• get a total % for the course – still looks equidiscriminating• convert to letter grade (A, B, C, D, F – ordered categories) or pass/no pass (classification)• letter grade is converted to number (A = 4.0, B = 3.0, -- back to equidiscriminating, but with fewer different values)• averaged across classes to get GPA – looks equidiscriminating• divide people up based on GPA – summa cum laude, magna cum lade, cum laude… (back to ordered categories)The “type” of scale or measurement you are trying to make is important, because it will influence the items you want for it …• a test to “identify remedial math students” will emphasize items that most students at that grade level can answer• a test to “identify gifted math students” will emphasize items that very few students at that grade level can answer• a test to “measure mathematical ability” will include items with a broad range of difficulty – so that we can place all students along the underlying continuum these “depression scales” will all have different “levels” and “ranges” of items…• identify clinically depressed individuals• research outcome variable for treatment of clinically depressed individuals• identify college students with little or no depression• research measure to examine relationship between depression and school performanceYou also have to pay attention to this when selecting scales for research – is the scale designed to give you the kind of measure you want (equidiscriminating vs. classificatory) for your target population????Desirable Properties of Psychological MeasuresInterpretability of Individual’s and Group’s ScoresPopulation Norms (Typical Scores)Validity (Consistent Accuracy)Reliability (Consistency)Standardization (Administration & Scoring)Standardization• Administration -- “given” the same way every time• who administers the instrument•