L9 Analyzing and presenting quantitative data

Question 1

What is the null hypothesis and the alternative hypothesis?

Answer 1

Every statistical test begins with a null hypothesis. The test assesses the possibility of an alternative hypothesis.

Ho = No difference between the groups; both samples come from the same underlying population

Ha = There is a difference between groups; samples come from different underlying populations

Question 2

What are the classifications of quantitative data?

Answer 2

Numerical (ratio or interval) and categorical (ordinal and nominal)

Question 3

How do you prepare your quantitative data for analysis?

Answer 3

• Entering your scores (use descriptive statistics to check the accuracy, code your categories)
• Missing values? (Why are they missing; random / not random? Code in SPSS with number that you can’t interpret as data)
• Calculate total scores

Question 4

What assumptions do your data need to meet?

Answer 4

• Outliers–> should be removed
• Linearity–> relation between dependent and independent variable should be linear
• Normality–> your data should fit into a bell-curve (normal distribution)
• Homoscedasticity–> equal variances

Question 5

What should you take into account when presenting your data?

For what classes of quantitative data would you use bargraphs, histograms, line graphs, and pie charts?

Answer 5

• If you use graphs, it should add something to the text
• Choose a chart that est fits your data type
• Keep it simple
• Color vs grey scale (printing)
• Don’t make people’s head tilt
• Order data according to logical hierarchy

Question 6

What are the 4 main descriptive statistics?

Answer 6

• Frequency: the number of instances in a group
• Central tendency: mean, median, mode
• Disperion (spread):
  
  • Range, interquartile range (Q3-Q1)
  • Variance: average of squared deviations of
    individual scores from the mean
  • Standard deviation: square root of the variance
  • Z-score: how many SD’s below or above the
    population mean a raw score is
    > Z-table = relation between probability and
    standard normal distribution
• Normal distribution: mean = median = mode

Question 7

What is inferential statistics?

Answer 7

Making inferences about populations using sample data drawn from the population

Question 8

What is the difference between one-tailed and two-tailed tests?

Answer 8

One-tailed = assigns direction (better, higher, lower, etc.)

Two-tailed: without direction (they do / don’t differ)

Question 9

What is a 95% confidence interval, and what is the formula?

Answer 9

95% confident that the population mean falls within this range.

95% CI = X +- 1,96 x SEM

Question 10

What two types of errors exist?

Answer 10

Type 1 = False positive–> shows positive, is negative

TYpe 2 = False negative–> shows negative, is positive

Question 11

What is the RR?

Answer 11

Risk ratio / relative risk

• Prospective study design
• Based on incidence measures
• “Group A has a x times higher risk on y than B”

Question 12

What is the OR?

Answer 12

Odds ratio

• All study designs
• Based on odds (the ratio between having / not having the outcome)
• “Group A has a x times higher odds on y than B”

Question 13

When do you use a one-sample t-test?

Answer 13

• 2 independent samples–> each person has been measured once
• Comparison of group of individuals
• Dependent variable on continuous scale

Question 14

When do you use a paired samples t-test?

Answer 14

• One sample–> each sample is tested twice; change over time
• Comparison within the same group
• Dependent varible on continuous scale

Question 15

When do you use ANOVA?

Answer 15

• Difference between 2 or more groups
• Ho = all population means are equal
• Ha = at least one population mean is different

Question 16

When do you use a chi-square test?

Answer 16

• To investigate whether distributions of categorical variables differ from one another
• Compares the tallies or counts of categorical responses between two (or more) independent groups