WEEK 5

Question 1

Levels of Data Measurement

Answer 1

-Data
-Categorical VS Quantifiable
- Nominal + Ordinal
VS Interval + Ratio

Question 2

Nominal Data

Answer 2

-Refers to Categorical Data
-Gender
-Ethnicity
-Job Type

Numbers are given to distinguish between categories, to rank them. e.g. 1= Female, 2= Male

Question 3

Ordinal Data

Answer 3

• Using a scale to put people in some sort of order/ rank, such as race positions
  -Unlike nominal, the size of the number does represent something

Question 4

Interval Data

Answer 4

• Put scores in an order, however the numbers are equal intervals
  -Temperature: Centigrade or Fahrenheit
• There is no absolute 0 where the variable being measured doesn’t exist, 0 degrees does not equate to 0

Question 5

Ratio Data

Answer 5

• The same features as Interval data: the differences between numbers are equal but there is an absolute 0
• e.g. height, scores on an achievement test, speed of a car

Question 6

Two types of statistics:

Answer 6

• Descriptive VS Inferential
  1. Summarisees data using numbers or graphs
  -Used to summarise all levels of data
  -Allows comparison across studies
  2. Use what we know from the data we have collected to make inferences and generalisations to the wider population

Question 7

The mean:

Answer 7

• definition: sum of all the scores divided by the number of scores in the sample
  -most commonly reported
  -most appropriate for ‘normal’ data

Question 8

The Median:

Answer 8

• Definition: The middle score/ value once all the scores in the sample have been put in rank order
  -Less commonly reported than the mean
• Organised form smallest to largest and then the middle score is selected

Question 9

The Mode:

Answer 9

• Definition: The most frequently occurring score/ category of scores
• Least commonly reported, useful for categorical variables
• Bimodal- 2 modes
  -Multimodal- Several modes

Question 10

Pros and Cons of each:

Answer 10

Mean:
+ Ease of calculation
+A good estimate of the population mean
+Ideal basis for inferential statistics
- Sensitive to extreme scores
-Cant be used for nominal data

Median:
+ Not sensitive to extreme scores
+Only requires ordinal levels of data
- Cannot be used for nominal data
- Not ideal as a basis for inferential statistics

Mode:
+ Can be used with any type of data
-May not represent central tendency at al if the distribution is skewed
-Not ideal as a basis for inferential statistics

Question 11

The population mean and Sampling Error

Answer 11

-The typical score in a population: Population mean

Sampling Error: the difference between the sample statistics and the population statistic

-the larger the sample, the closer the sample mean to the population mean

Question 12

Graphical Descriptions of Data

Answer 12

-Bar Chart: used to summarise a categorical variable

Question 13

Measures of variability

Answer 13

1. The range
2. The interquartile Range
3. The standard deviation

Question 14

The range

Answer 14

• Definition: The distance between the lowest and highest score in a sample
  -Subtract the bottom value from the top value
• Sensitive to outliers

Question 15

The Interquaratile Range

Answer 15

• Definition: Distance between the upper and lower quartile in a set of data

-Appropriate for ordinal level data
-Appropriate for non-normal data
-Less affected by outliers

Question 16

The Standard Deviation

Answer 16

• definition: an estimate of the average deviation of the scores from the mean

Two ways to calculate:
- Corrected: used to estimate population standard deviation, SPSS uses this formula ‘
-Uncorrected: Used when you are not using the standard deviation to make estimates of the underlying population