Types of variables and levels of measurement

Variables Basics

variables are identified events that change in value
many explanatory concepts in psychology are unobservable directly but are treated as hypothetical constructs
Variables to be measured need precise operational definitions: the steps taken to measure the phenomenon and/or set of activities required to measure X
Independent variables are assumed to affect the dependent variable, especially if they are controlled in experiments
confounding occurs when a variable related to the independent variable obscures a real effect or produces the false impression that the independent variable is producing observable changes

Independent and Dependent variables

dependent variable: variable which is assumed to be directly affected by changes in the independent variable

independent variable: manipulated variable in an experiment which is assumed to have a direct effect on the dependent variable

Example:

participants are given a list of words to learn under 2 conditions - 30s or 1m.
independent variable is the time given to learn words
dependent variable is the number of words that were recalled.

Extraneous variables

an extraneous variable: any variable other than the independent variable which might have an effect on the dependent variable

most often used in experiments where there is interest in controlling the unwanted effects of all variables except the independent variable

Random variable/error: unpredictable and random variables that do not impact one condition more than another, e.g., feelings on the day tested, atmosphere of the room, temperature of the room.

constant variable/error: variables which systematically impact one condition more than another, e.g., more interesting instructions, easier list of words, more practice

confounding variables: variable which is uncontrolled annd obscures any effect sought, usually in a systemic manner

Measurement Basics

The levels at which data can be measured are: nominal, ordinal, interval and ratio

Nominal data: simple classification
Ordinal data: cases are ranked or ordered
Interval data: intervals equal in amount
Ratio data: interval but include a real 0 and relative proportions

Attempts are made to convert many psychological scales to interval level using standardisation

All variables can be classified according to whether they are categorical or measured

Measured variables may be measured on a discrete or continuous scale

Nominal Level of Measurement

Categories and you do not need to count in order to distinguish one item from another
can only be discrete
only the mode can be used
Example - colour of horses, sex,

Ordinal level of measurement

ordinal numbers represent the position in a group but not the distance between positions
generally 0.5 as the smaller unit = can only be discrete.
median is the most appropriate measure of central tendency (but mode can be used)
Example - Year group, position in race

Interval level of measurement

distances between points on the scale are all equal for equal units
can be either continuous or discrete
mean is the most sensitive measure of central tendency (but mode and median may be used)
Example - temperature, grades, iq

Ratio level of measurement

real 0 and proportionality across the whole scale. no negative numbers.
can be either continuous or discrete
Example - race times, number of words. remembered

Continuous Scale of measurement

no limit to the subdivisions of points which can occur
Example - time, height

Discrete scale of measurement

each point on the scale is entirely separate from the next
Example - number of children, number of words