Research methods in practice (wk8)

Question 1

What are measures of movement?

Answer 1

Gait (walking), biomechanics, posture and balance

Question 2

How to measure physical activity

Answer 2

• Self-report measures -> e.g. Global PAW, GP-PAQ. Open to bias and over estimation. Vague time periods.
• Pedometers -> Popular for low-cost, large sample activity monitoring. Simple operation (battery). Only 1-dimension of data- daily step count.
• Research grade activity monitors -> Ranging from low-cost e.g. axivity (simple functions, for large samples). Too high-cost e.g. Actigraph, APDM (multiple functions)

Question 3

What are the sensors i consumer devices and what do they measure?

Answer 3

-Inertial measurement unit (IMU)
-Which measure acceleration and orientation

Question 4

What can we measure while wearing devices?

Answer 4

• Frequency -> step count, number of bouts
• Intensity -> cadence, metabolic equivalence of task (MET)
• Time -> Periods of moderate-vigorous PA (MVPA), sedentary periods
• Type -> Inclines/ stair climbing, walking/ running, sitting/standing

Question 5

How can we make use of this data for understanding health/lifestyle/ behaviour?

Answer 5

• Randomised Controlled Trial (RCT): Comparing a placebo or healthy group with health condition and intervention
• Pre-Post Test: Capturing a baseline prior to intervention
• Combining with questionnaire data: Allow grouping or predictors of change
• Combining with qualitative data: Detailed individual insights
• Identifying patterns using data-driven analyses (e.g. machine learning)

Question 6

What are the 2 types of anonymous data?

Answer 6

• Anonymised data -> Remove all identifiable data (e.g. name, address, mobile number). No way of tracking data back to an individual
• Pseudonymised data -> As above, but data has a participant ID. Separate file contains ID and identifiable variables. People with access to file can track back to an individual.

Question 7

What is the ‘Motivated Intruder Test’?

Answer 7

-Adding extra layers of privacy -> Keeping detailed variables – always a chance an adversary will be able to de-anonymise the data. Removing all of the signal in a dataset to anonymize, can make some analyses pointless.

Question 8

What is differential privacy?

Answer 8

1. Insert random noise into the information made available
2. Done correctly, meaningful answers can still be retrieved

Question 9

What is the method which can be used to combine objective and subjective data to add context to objective measures?

Answer 9

-Combine objective (e.g. daily step count) with subjective (e.g. questionnaire) date to get the context
1. Objective data variable, e.g. change in step-count: Dependent variable
2. Questionnaire variables, e.g. age, gender: Independent variables