Introduction to quantitative research designs

Question 1

What is a research design?

Answer 1

a framework or blueprint for conducting the research project. It details the procedures that are necessary for obtaining the right information needed to structure and/or solve the research problem”

Question 2

Why does research design matter?

Answer 2

• Efficiency-value for money
• It’s a blueprint for advance planning
• Error reduction
• Reliability and evidence design dependent
• Is the design ethical
• Allows hypotheses to be tested

Question 3

What questions need to be asked in terms of internal validity?

Answer 3

• Was your study done correctly?
• Can you claim to have cause and effect relationships?
• Have you controlled for other causes?
• Are your measures appropriate and valid?

Question 4

What questions need to be asked in terms of external validity?

Answer 4

• Is your study population representative?
• Did lots of people drop out?
• Can you make generalisations?
• Selection bias?

Question 5

What are the two common research designs?

Answer 5

• Observational/correlational

* Experimental

Question 6

What are types of observational/correlational designs are there?

Answer 6

• Case study
• Case series
• Cross sectional study
• Case‐control study
• Cohort study

Question 7

What types of experimental designs are there?

Answer 7

•Before and after study-(Quasi experimental)
• Interrupted time series (Quasi experimental)
• Controlled study
• Randomised controlled
trial
• Systematic reviews of RCTs

Question 8

Explain correlational designs

Answer 8

• Look at the impact of a variable (independent) and relationship with another variable (dependent)
• Participants choose independent variable
• Observing and recording only
• No intervention • Ethically neutral • Correlation not causation

Question 9

Explain case study designs

Answer 9

• Sometimes called case report
• Usually a single case
• Interesting finding
• Limited generalisability
• Very limited causality
• Hypothesis generating
• Retrospective

Question 10

Explain cross-sectional design

Answer 10

• Data collected on series of patients (participants)
• Single time point
• What is happening now
• Associations between variables only not causality
e.g. Youth Sport Survey

Question 11

Explain a case-control study

Answer 11

• Outcome precedes exposure
• Go back in history to try and understand how some exposure linked to disease risk
• Good for rare outcomes
• ? Recall bias

Question 12

Explain a cohort study

Answer 12

• Also called longitudinal study
• Follow a population over time
• Uses questionnaires, interviews, etc.
• Temporal relationship between variables
• Exposure precedes outcome
• Won’t measure every outcome or exposure
• Collection of data on a series of variables at multiple time points

Question 13

What are the common threats with correlational designs?

Answer 13

• Recall bias
• Difficult to get random samples
• Cannot account for all confounders
• Cannot prove causation only correlation
• Temporal relationship not always clear
• Attrition bias e..g people dropping out

Question 14

Explain experimental design?

Answer 14

• Widely used
• Manipulates one variable called the Independent
Variable (IV) to see what effect this has upon another variable called the Dependent Variable (DV)

Question 15

Explain before and after study

Answer 15

• Single group measured before, then intervention, then measured again
• Learning effects
• Easier with smaller samples
• Cheap and easy
• Could they just improve (get fitter) anyway?
• Unmeasured confounding?

Question 16

Explain an interrupted time series study

Answer 16

• Single group measured before, then intervention, then measured again
• But with multiple measurements before and after
• Can examine seasonal trends
• Useful for examining policy impact
• More robust to bias then before‐after studies
• Cheap if have access to routine data

Question 17

Explain Randomised Control Trials

Answer 17

• Participants randomly allocated intervention or control
• Prospective
• Gold standard
• Can determine cause and effect
• BUT
• Expensive
• Not always ethical

Question 18

What is the structure of experimental data?

Answer 18

• Obs ‐ An observation made in relation to the dependent variable; there may be two or more observations, before (pre‐test) and after (post‐test), the experimental manipulation
• Exp – The experimental intervention (manipulation of the independent variable).
• No exp refers to the absence of an experimental intervention and represents the experience
of the control group
• T – The timing of the observations are made in relation to the dependent variable.
• The difference between each group’s pre and post test scores is then analysed to establish whether or not Exp has made a difference

Question 19

What is the criteria for causality?

Answer 19

• Theoretical plausibility: Does theory/biological understanding support this inference
• Co‐variation: As independent variable changes so does the dependent variable
• Time order: The exposure (independent variable) happened before the outcome (dependent variable)
• Non‐spuriousness: Have other variables been accounted for