Measurement Scales & Data Characteristics

Question 1

Types of Data

Variable, Data, and Factor

Answer 1

Variable - question or field name

Data - answer or observations about the variable

Factor - variable of interest in an analysis

Question 2

4 Basic Measurement Scales

Answer 2

1. Ratio
2. Interval
3. Ordinal
4. Nominal

Question 3

Measurement Scale Groupings

Quantitative/Qualitative

Answer 3

1. Quantitative - numerical
   
   1. Discrete - whole numbers
   2. Continuous - decimals
2. Qualitative - categorical

Question 4

Cardinal and Count Data

Answer 4

Quantitative and discrete

Question 5

Quantitative Continuous Scales

Answer 5

Interval and Ratio

Ratio -

absolute “zero” treated as the point of origin, and equal + definitive ratio between sets

can be multiplied

positive

Interval-

No absolute zero (temperature - Kelvin)

can be negative

Question 6

Time Data Scale

Answer 6

• Continuous
• keeps restarting from zero at set periodicity
• distance b/w readings is measurable and comparable
• Need to consider seasons/time-zones

Question 7

Qualitative Scales

Answer 7

1. Ordinal
   
   • ​ordered​
   • distance b/w points not known or varies
   • no absolute “zero”
   • no math should be done on results
2. Nominal
   
   • ​subgroups - with or without order and dichotomous​

Question 8

One hot encoding

Answer 8

Making qualitative data into 1s and 0s

Question 9

New Data Types

Answer 9

1. Images
2. Video
3. Audio
4. GPS data
5. Unstructured text

Question 10

What affects data reliability?

NSFTEL

Answer 10

Data characteristics:

1. Nature
2. Source
3. Format
4. Timing
5. Extent
6. Level of aggregation

Question 11

What is the importance of the nature of data?

Answer 11

• Financial vs nonfinancial
• Quantitative vs qualitative
• Completeness
• Accuracy
• Fitness for purpose

Question 12

Reliability

Answer 12

Consistent and repeatable

Question 13

Valid data

Answer 13

measures what you think it is measuring

Question 14

Data quality

Answer 14

Data quality = reliability + validity

Question 15

Data integrity as a subset of data quality

Characteristics

Answer 15

1. Completeness
2. Uniqueness
3. consistency
4. timeliness
5. validity
6. accuracy

Question 16

data range

Answer 16

data max - data min

outliers heavily affect this

Question 17

Before you begin your analysis, establish estimates of reliability for your chosen variables. It’s a mark of statistical due diligence, and your clients will thank you for it.