Word Doc Week 2

Question 1

What should I do before conducting any data analysis?

Answer 1

• “Screen & Clean” Data
• SPSS can not tell if data is ridiculous
• Only the researcher knows this so I need to make sure I screen the data for errors

Question 2

Six reasons for conducting exploratory data analysis

Answer 2

1. Checking for data entry errors
2. Obtaining a thorough descriptive analysis of your data.
3. Examining patterns that are not otherwise obvious
4. Analysing and dealing with missing data
5. Checking for outliers
6. Checking assumptions

Question 3

EXPLORE command in SPSS

Answer 3

• The explore command is SPSS covers all bases.

Question 4

Who is John Tukey

Answer 4

• A very practical statistician
• Language and techniques of EDA developed by him

Question 5

What does Screening and Cleaning involve

Answer 5

• Computing new variables from existing ones
• Recoding variables
• Dealing with missing data

Question 6

Checking Data Entry Errors

Answer 6

• Use the frequencies command to check for data entry errors in categorical/nominal variables
• Use the outliers option in the explore command to check for data entry erros in continuous/scale variables

Question 7

Options for Dealing with Data Entry Errors

Answer 7

1. Remove data
2. Make ‘educated guesses’ about what was intended

Question 8

Obtaining a thorough descriptive analysis of data

Answer 8

• The explore command provides more descriptive stats than any other procedure
  
  • Multiple measures of central tendency
  • Multiple measures of variability
  • Quantitative measures of shape
  • Confidence intervals
  • Percentiles

Question 9

Examining Patterns that are not otherwise obvious

Answer 9

• Stem and Leaf Plots
• Box and Whisker Plots

Question 10

Analysing and dealing with missing data

Answer 10

Do you leave it out or do you substitute in the mean?

Question 11

Outliers

Answer 11

• Are they legitimate, or are they error?
• Do you keep them? Do you discard them?
• Balancing act, as with so much in data analysis

Question 12

Assumptions

Answer 12

• All parametric procedures (e.g., t-tests, ANOVA’s, correlation) operate under certain assumptions
• Main two:
  
  • Normality
    
    • Assumes that your data comes from population that is normally distributed
  • Homogeneity of variance
  • Assumes that, if your data is to be divided into groups, the level of variability in the groups will be approximately equal (i.e., not significantly different).

Question 13

Normality is tested in four ways:

Answer 13

1. Visual inspection of histograms and stem and leaf plots
2. Visual inspection of normality and detrended normality plots
3. Normality tests
4. Skewness divided by SE skewness

Question 14

Subjective Normality Tests

Answer 14

1. Visual inspection of histograms and stem and leaf plots
2. Visual inspection of normality and detrended normality plots
   
   • not influenced by sample size

Question 15

Objective Normality Tests

Answer 15

1. Normality tests
2. Skewness divided by SE skewness

• Objective but influenced by sample size
• With a large sample, even trivial deviations from normality will indicate a violation of the assumption

Question 16

The least important assumption

Answer 16

• Normality
• Statistical tests are very robust to even gross deviations from normality, particular when you have a large sample

Question 17

Why is normality the LEAST important assumption?

Answer 17

• Statistical tests are based on sampling distributions
• Sample distrbutions will always be normal regarles of the underlying population distribution
• they will get more normal as the sample size increases

Question 18

When is normality a problem?

Answer 18

• When the deviation is strong and the sample is small

Question 19

Tests 3 and 4 are sensitive to sample size

Answer 19

• Ironical
• Indicate a violation of normality with a large sample
• Even if deviation is trivial
• Normality is rarely a problem when the sample size is large

Question 20

Transformation

Answer 20

• Applying a simple mathematical operation to the data to deal with violations of assumptions

Question 21

Common Transformations

Answer 21

1. Log 10
2. Square root
3. Reciprocal

Question 22

Homogeneity of variance assumption

Answer 22

• More serious particularly when the sample is unbalanced
• tested using dedicated test: Levene’s Test
• Dealt with by power transformation

Question 23

Levene’s Test

Answer 23

• Used to test if k samples have equal variances.
• Equal variances across samples is called homogeneity of variance.
• Some statistical tests, for example the analysis of variance, assume that variances are equal across groups or samples.
• The Levene test can be used to verify that assumption.

Question 24

Some general comments about transformations

Answer 24

• They are no “magic bullet”.
• They can’t cope with zeroes (need to add a constant)
• They are unpredictable (e.g., a transformation designed to deal with normality can fix up homogeneity of variance and vice versa)
• Some data is “untransformable”

Question 25

Recoding Data:

Answer 25

• Often need to recode for a whole host of reasons:
  
  1. Reducing numbers of groups
  2. Reverse scoring