# PSYB01H3 Lecture Notes - Parametric Statistics, Chi-Squared Test, Nonparametric Statistics

18 views5 pages

Chapter 4- Conceptualization and Measurement

Measuring Mental Health and Mental Illness

•being mentally healthy means u dunt have a diagnosable mental disorder or

disability

ohowever u may say it’s a mentally happy person is happy mm thriving

ect

oso how do u no which one it is and how develop constructs

•Case study: Keyes (1997) - Is mental health than just lack of mental illness?↑

oRooted in positive psyc- argues that thers to mental health then simply↑

the absence of negatives

oLearn about the construct on the next slide cuz that’s wat the point

basically is

Developing and Defining Constructs

•define ur ideas into a clearly defined theory- which provides the conceptual

framework for ur studies, meaning of terms and identifies how variable re to

measured and the rationale

Theoretical Constructs

•theory unites concepts with the variables and measurements and it does this

by translating abstract concepts into a calculable variables which all have

objective measurement

•abstract concepts are referred to as constructs

•construct – specified in a theory in order to describe , explain and predict a

wide range of related empirical findings

•Not directly observable →Must be translated into an operational defn

Operational Defn of Constructs

•operational defn-provides a recipe for spec variables that are to be used to

measure a construct

ooperational defn of a construct isn’t limited to spec how to measure it

oin lab studies it can spell out conditions under which the construct is

expected to occur – it can identify the independent variables that are

expected to prod the construct

•eg construct of anxiety can be in 3 distinct variables—measures of

psycphysiology, obs ratings and self report →rep 3 diff dependent variables of

the same construct,

•if u have time read pgs 97-98

•structure of a scientific theory begins with abstract constructs which in turn are

translated or operationalized into measurable independent and dependent

variables

•sound prog of research depends on the extent to which these operations

capture the nature of the theoretical constructs under investigation

•scientific validity of a theory or a theoretical construct hinges on the extent to

which it descries, predicts and explains those empirical relationships that are

uncovered between independent and dependent variables

•many times the relationships between independent and dependent variables

will be linked , connected or mediated by intervening variables , hyp internal

states that are critical components of deductive struct of a theory

Building Constructs

•mental health vs mental illness ... keys adopted a conceptual framework

rooted in the school of psyc known as positive psyc

•positive psyc- scientific study of ordinary human strengths and virtues seeks

to learn about human thriving , optimism , joy, capabilities

•pgs 100-101

Measurement of Psyc Constructs

•We investigate if our hyp constructs reflect objective reality through

measurement?

oanswer is to find the correct objective reality we use a lot of methods

•measurement - Assignment of numerals to variables according to some

objective rule

oSo then u can use #s instead of just names

•use multiple methods of measurement is essential

oessential for use to develop a complete understanding of a phenomenon

and its relationships

odiif forms of measurement for the same construct allow understanding ↑

Levels of measurement

•measurement : process by which #s are used to designate objects or events

according to some objective rule

•rule- means the way #s are used to rep a particular scale of measurement

ocan have 1 of 4 scales which range in complexity →nominal( least

complex) , ordinal, interval and ratio( most complex)

•nominal ( also called categorical or qualitative level) – identifies variables

whose values have no mathematical interpretation→ they vary in kind or

quality but not amount

ovariables values are mutually exclusive- of every case can only have one

value

ovariables values are exhaustive when every case can be classified into

one of the categories

owhen variables are mutually exclusive and 0065haustive- every case

corresponds to 1 attribute ( value )

•→Special case of dichotomies- variables have 2 values but still measured at

the nominal level

•ordinal level of measurement – 1st of 3 quantitative labels – the #s assigned

to response choices allow for greater than and less than distinctions

ocuz it allows for ranking it’s also referred to as ranking scale

oDoes not assume equal distance between values

olikert scale is common metric system , 1- none of the time , 2 a little of

the time , 3 some of the time , 4 most of the time , 5 all of the time

•interval – has all characteristics of nominal and ordinal scales of measurement

oit gives a name or category for each obs with the # serving as a code for

the able

olike ordinal the responses are numerically ordered or ranked form lowest

to est on the same characteristic↑

oEqual distances between values and No true zero point

ou can’t say 80 C is twice as much as 40

•Ratio level – the #s rep fixed measuring units and an absolute zero point ( 0

means absolutely no amount of whatever the variable indicates )

## Document Summary

Developing and defining constructs: define ur ideas into a clearly defined theory- which provides the conceptual framework for ur studies, meaning of terms and identifies how variable re to measured and the rationale. Important when considering statistical analyses: non-parametric statistics = test hyp for variables that use nominal and ordinal scales, eg chi square test, parametric statistics = generally used with variables that are measured on either interval and ratio scales. Measurement reliability: test-retest reliability= scores should remain consistent over time. Interobserver reliability= degree to which raters code the same observed behavior. Logic of measurement validity: circular reasoning use of independent measures of theoretically liked outcomes to validate the assessment protocol effectively avoids a common pitfall in research design, research design can be invalidated by circular reasoning. Content validity: valid test needs to provide full measure of a construct, content validity- refers to extent to which the test items have sufficient breadth to capture the full range of a construct.