Descriptive Statistics

Question 1

Name 3 types of data
(CDC)

Answer 1

Categorical

Discrete

Continuous

Question 2

Categorical
Binary or Nominal

Answer 2

Has two or more categories with no ordering to them.

E.g. Hair colour, Job title

Question 3

Continuous
Ratio or Interval variables

Answer 3

Can take any fractional value
E.g. Reaction times

Question 3

Discrete
Ordinal, Ratio, or Interval variables

Answer 3

Has a fixed value with a logical order
E.g. Shoe size, Score out of 10

Question 4

Median Cons?

Answer 4

Ignores a lot of the data

Difficult to calculate without a computer

Can’t use this with NOMINAL data

Question 4

Median Pros?

Answer 4

Insensitive to outliers

Often gives a real, meaningful data value

Useful for ordinal data, and skewed interval/ratio data

Question 4

What are the 3 measures of central tendency?

Answer 4

Mean- sum of data points
Median- middle score in data set
Mode- most in a data set

Question 5

Median:
The middle value in a dataset, or the mean of the middle two values can be calculated as:

Answer 5

Odd value datasets: (n+1)÷2

Even value datasets: Line up middle two values then÷2

Question 5

What is the equation for calculating the mean?

Answer 5

Sum of individual data points
÷
sample size

Question 6

Mean Pros?

Answer 6

Uses all of the data

Is most effective for normally distributed datasets

Question 7

Mean Cons?

Answer 7

Sensitive to outliers

Values are not always meaningful (we cant get a score of 6.74 out of 10!).

Only meaningful for RATIO and INTERVAL data

Question 7

Measures of spread:
Mode

Answer 7

no measures of spread

Question 8

Measures of spread:
Median

Answer 8

‘distance-based’ measures such as range and interquartile range

Question 9

Measures of spread:
Mean

Answer 9

‘centre-based’ measures of spread such as variance and standard deviation

Question 10

Interquartile range IQR:

Pros and cons are identical to the median

Answer 10

highest score - lowest score
but ignores most extreme values

Lower quartile= median of lower half of the data

Upper quartile= median of upper half of the data

Interquartile range = Upper quartile-Lower quartile

Question 10

IQR example:

Answer 10

Lower quartile =14th/15th value
=5 (Five times a day)

Upper quartile = 43rd/44th value
= 7 (Seven times a day)

IQR = 7-5 = 2

Question 10

Variance Pros?

Answer 10

Uses all of the data

Forms the basis of several other tests

Question 10

Deviance-

Answer 10

take each score and subtract it from the mean

Question 10

Variance Cons?

Answer 10

Requires a normal distribution

Sensitive to outliers

Units are not sensible (can we explain variance as scores2?)

Question 10

Sum of squared errors-

Answer 10

total the squared errors

Question 10

Squared errors-

Answer 10

take each deviance score and square it

Question 10

Variance-

Answer 10

average squared errors

Question 10

What is measure of spread that is equal to the unit of measurement of the dependent variable.

Answer 10

Standard Deviation (SD)

Question 11

How is SD calculated?

Answer 11

Calculated using the square root of variance.

Question 11

What does Ordinal data mostly use?

Answer 11

Median and IQR

Question 11

What does Categorical data mostly use?

Answer 11

Mode

Question 12

lmao you should literally know all of this from a level!

Answer 12

FRFR