Professional Documents
Culture Documents
➢ Scales of Measurement
■ Four types of scales exist:
● (a) nominal: numbers are assigned to discrete labels or categories
(e.g., race, gender, college major)
◆ LEAST Precise
◆ Simplest form of measurement
● (b) ordinal: attributes are ranked in ascending or descending order
(e.g., best to worst performance)
● (c) interval: zero point is arbitrary but distance between scores
has meaning (e.g., intelligence or interview scores)
● (d) ratio: distance between scores has meaning and there is a true
zero point (e.g., salary, typing speed)
◆ MOST Precise
■ The degree of precision with which we can measure differences among
people increases as we move from nominal to ratio scales.
● Least Precise → Most Precise
■ Precision is a function of the scale of measurement
❖ Scales of Measurement
➢ Nominal Scales & Ordinal Scales
■ A nominal scale is composed of two or more mutually exclusive
categories. Examples include:
● Applicant Gender:
◆ 1. Male
◆ 2. Female
● Job Title:
◆ 1. Sales Manager
◆ 2. Sales Representative
◆ 3. Sales Clerk
■ An ordinal scale is one that rank-orders objects of individuals from high
to low
■ Example of an Ordinal Scale
● Supervisory ranking of subordinates
◆ Problem: Don;t have info on the magnitude of the
differences between two
● This only shows that One is greater than Two
● Only tells us who is greater than and who is less than
■ Example of an Interval Scale
● We want to use this in the HR world; You will see this in HR
● Tells us what the magnitude of the difference between the ranks
are; more information
➢ Ratio Scales
■ A ratio scale has an absolute zero point, and differences between
numerical values have meaning
● Example: most scales involving physical measurement or counting
■ MOST PRECISE AND MOST SOPHISTICATED
■ WILL NOT SEE THIS USED IN HIRING PROCESS
■ HAS TO DO WITH PHYSICAL
❖ Levels of Measurement
➢ Summary of Primary Scales
❖ Standardization of Selection Measurement
➢ Standardization: a means of controlling the influence of outside or unimportant
factors on the scores generated by the measure
■ Differences in scores must be attributable to ability (true differences), and
NOT to other factors
➢ A predictor or criterion is standardized if:
➢ 1. Content. All people measured by the same content
➢ 2. Administration. Information is collected the same way in all locations
● UNLESS ADA APPLIES
➢ 3. Scoring. Rules for scoring exist and are applied in the same way
➢ Assessment/Selection devices often fail to meet the requirements of
standardization
➢ We need to always ask the same questions for every candidate in interviews
❖ Standardization
➢ Predictors and Criterion Measures
■ Predictors or Selection Procedures
● Many types have been used, but most fall into three broad
categories:
◆ 1. Background information
◆ 2. Interviews
◆ 3. Tests
➢ Ability Tests and Aptitude Tests
■ Criteria or Measures of Job Success
● One way to classify criteria is by the measurement method used
to collect data:
◆ 1. Objective Production Data
➢ How many goods do they produce in a day
◆ 2. Personnel Data
➢ Absenteeism,
◆ 3. Judgmental Data
◆ 4. Job or Work Sample Data *
➢ Criterion Variable: “Lifting 40lbs”
◆ 5. Training Proficiency Data
➢ Are you trainable?
➢ Finding & Constructing Selection Measures
■ Once we have identified the criteria necessary for successful work
performance, the process of identifying and implementing our selection
procedures may begin
● A consultant, perhaps an industrial-organizational psychologist,
may be needed
■ Two choices:
● (a) locate and choose from existing measures
● (b) construct our own
➢ Using Percentiles
■ MOST COMMON NORM/ Most frequently used tool
■ Ex. If you make 75 on the test, what percentile are you in?
■ Percentiles scores are used to show the percentage of persons in a
norm group who fall below a given score on a measure
● A percentile score is not a percentage score
➢ The higher the percentile score, the better a person’s performance
relative to others
➢ Percentile scores useful, but subject to misuse