Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Principles of Research > L18 - Data Acquisition > Flashcards

L18 - Data Acquisition Flashcards

1
Q

Distinguish between 1º data and 2º data

A
  • 1º Data collected specifically by investigator to answer resh qn
  • 2º Data: existing data collected for other purpose by someone else, which an investigator can use to answer resh qn
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Compare 1º and 2º Data

A

1º Data VS 2ºData:

  • Collection purpose: Specific to resh VS other purpose
  • Variables: Specific variables to resh VS whatever colelcted
  • Missing data: can be minimised VS cannto control
  • Accessible ppn: Less representative VS more representative of ppn
  • Process: Less efficient vs More efficient
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

List the three broad methods of primary data collection

A
  1. Experimental method: Lab and Controlled trial
  2. Observation Method
  3. Survey Method
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Consideration when using experimental lab method to collect primary data

A

REPRODUCIBILITY

- If I repeat expt, can I get similar results?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Describe the observation method used to collect primary data

A

Resh method that enables researchers to systematically observe and record people’s behaviour, actions and interactions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

A study observes the helmet wearing habits of motorcycle riders in country X by employing a few investigators in different parts of the country to observe and count the number of riders wearing helmet

State the type of primary data collection method (and subtype if possible)

A

Non-participant Observation method

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

A study aims to find out the effect on intensity of light on the productivity of factory workers. The investigators join the factory workers for one month and the light intensities are varied. The investigator collects data on the number of unit of products per hour for different intensities of light.

State the type of primary data collection method (and subtype if possible)

A

Participant observation method (since researcher is actively involved with subjects)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Strengths of Observation method to primary data collection

A
  1. Independent of subjects’ willingness to respond
  2. Reduce reporting bias (people cannot lie as researcher directly see what people do)
  3. Useful for subjects who are unable to provide verbal reports of their feelings/thoughts
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Limitations of Observation method

A
  1. Time consuming
  2. Researchers must refrain from interpretation, else may intro bias (hence rq skilled observers)
  3. Susceptible to Hawthorne effect: subjects aware of being observed and change their behaviour, resulting in inaccurate results
  4. Unable to establish cause of behaviour observed
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Some considerations of observation method

A
  1. What to observe?
  2. How to record observation? E.g. film, audio, pen and paper
  3. How to ensure accuracy of observation?
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Describe the survey method of collecting primary data. List the three broad survey method in data colelction

A

Survey: Getting insights on an issue from individuals or group of people with or without direction interactions

Methods:

  1. In-depth (one-to-one)
  2. Focus group discussion
  3. Questionnaire
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Strength and Limitations of in-depth interview to collect primary data

A

Strengths:

  1. Good for sensitive topics due to privacy
  2. Able to get deeper information
  3. Can get more info on personal experiences and feelings

Limitations:

  1. Less efficient (time, a lot of transcription)
  2. Interviewer bias
  3. Needs rapport and flexibility, which interviewer must be skilled
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What are some approaches in conducting an in-depth interview with individuals?

A
  1. Use semi-structured interview guide (for prompting while preserving flexibility)
  2. Establish rapport and motivate interviewee by probing
  3. Ask OPEN and EMPATHIC questions
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

List some considerations before conducting in-depth interview

A
  1. Require accurate translation to fit country culture?
  2. Pilot-testing?
  3. Data collection? (E.g. video, or audio only)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

List some considerations for focus group discussion in collecting primary data

A
  1. Homogeneity: participants more open if they have people similar to themselves
  2. Level of familiarity: strangers preferred for anonymity, allow participants to share discuss more openly and willingly (privacy ensured)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What should the discussion guide of a focus group discussion contain?

A
  1. Translation
  2. Pilot-testing
  3. Language and translation issues
  4. Data collection (similar to in-depth interview)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

Strengths of focus group discussion in collecting primary data

A
  1. A lot of info
  2. Wider variety of views
  3. May identify new issues
  4. Useful for resh with the nature to explore, explain and evaluate
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Limitations of focus group discussion method

A
  1. Interviewer bias
  2. May sidetrack (hence required skilled mod)
  3. Less confidential: participants may not want to share
  4. Social pressure to conform: response may hence not be independent
  5. Info may not be representative of wider ppn
  6. Info obtained is not valid at individual level (similar to ecologic study)
19
Q

Describe Questionnaire. What is it useful in

A

Number of questions printed or typed in definite order on form or set of forms with purpose to seek SPECIIFC info from respondants

Useful if:

  • Have limited resources
  • Get data on knowledge, beliefs and attitudes
  • When privacy is a concern
20
Q

Factors affecting quality of data collected by questionnaire

A
  1. Questionnaire construction

2. Modes of questionnaire administration

21
Q

What are the considerations of questionnaire construction?

A
  1. Structure
  2. Content
  3. Language
  4. Pre-testing
  5. Reliability and validity
22
Q

General structure of a questionnaire. A reason why its structured that way

A
  1. Intro
  2. Demographics
  3. Factual Qn
  4. Opinion Qn
  5. Closing and return instructions

Opinion after factual, coz opinion qn usually less comfortable

23
Q

Steps to constructing a questionnaire

A
  1. Think of info to be sought
  2. Draft questions
  3. Pre-test quesitonnaire
  4. Revise questionnaire if required
24
Q

List the type of questions that make up a questionnaire

A
  1. Close-ended (CEQ)

2. Open-ended (OEQ)

25
Q

What are the strengths and limitations of OEQ?

A

Strengths:

  1. Detailed
  2. Disallow successful guess
  3. Allow qualitative resh

Limitations:

  1. Interviewer administered: difficult to record systematically and difficult to probe without bias
  2. Self-administered: High non-response rate
  3. Hard to classify and code info for data analysis
26
Q

What are the strengths and limitations of CEQ?

A

Strengths:

  1. Standardised responses
  2. Easy to code and analyse data
  3. Less time rq than OEQ

Limitations:

  1. Little depth in answers
  2. Bias if answer options are incomplete
  3. Presentation format, and the nature of question imposing researchers’ priorities on respondants, causes responses to be affected
27
Q

List the types of CEQ and some examples where aplicable

A
  1. Checklists
  2. Rankings questions
  3. Rating scales
    - E.g. likert scales (Bipolar/unipolar), VAS
28
Q

What are things to avoid when formulating questions for questionnaire to collect primary data?

A
  1. Avoid double-barrel
  2. Avoid bias and leading questions
  3. Avoid technical jargons
  4. Avoid incomplete options (E.g. 1d,3d,5d but no even days)
29
Q

What to ensure during pre-testing of questionnaire

A
  1. Questions understood as expected
  2. Logical order of questions
  3. Okay duration
  4. Ensure qns help ans resh qn
30
Q

Distinguish between reliability vs Validity

A
  • Reliability: Reproducibility

- Validity: How well an instrument measures what it is intended to measure

31
Q

Advantage of self-administered vs interviewer administered questionnaire and their disadvantages

A

Self-administered:

  • Cheap
  • Less interviewer bias
  • Can mail/email

Interviewer-administered:

  • Higher response rate
  • More in-depth response
  • More control in filling questionnaire

Disadvantages, opposite of each others

32
Q

Formula for response rate

A

(No. of people who complete questionnaire / No. of people asked to do questionnaire) x 100%

33
Q

What is the issue of low response rate and its impact?

A

Main issue: Non-response bias as non-respondants are different from respondants

Effect:

  • Decreased internal validity (not valid for ppn in study)
  • Decreased external validity (only apply to participants who completed study, not generalisable)
34
Q

The formula for Yamane’s formula is:
n = N/(1+N(e^2))
Given p = 0.5, state what do each variable mean (including p)

A
  • n: Sample size required
  • N: Population
  • p: estimated proportion of an attribute present in population
  • e: desired level of precision
35
Q

Formula for Cochrane formula is:

  1. n0 = (Z^2xpxq)/e^2
  2. n = (n0)/(1+((n0-1)/N))

State what do each variable mean. State the purpose of formula 2 and when to use it

A
  • n0: sample size
  • Z: Z-score of normal distribution (1.96@95%CI or sig level of 0.05)
  • p: estimated proportion of an attribute present in population (if unknown, assume 0.5)
  • q = 1-p
  • e: desired level of precision

Usage of formula 2: when ppn is small such that n0 ≥ 0.05 ppn (more than 5% of population)
- Formula adjusts for small ppn

36
Q

List and describe RANDOM sampling methods for questionnaire administration

A
  1. Simple random sampling
    - All subject equal probability selected
  2. Systematic sampling
    - k = ppn/n, every kth subject selected
  3. Stratified random
    - Divide to strata
    - Take random sample for each strata
  4. Cluster random:
    - Divide to cluster
    - Randomly select ALL of ONE cluster

(SSSC)

37
Q

List and describe NON-RANDOM sampling methods for questionnaire administration. Explain the bias that might be introduced for each method

A
  1. Convenience
    - Anyone you meet
    - Biased: e.g. meet friendly people, or meet ppn that frequent a shopping mall
  2. Quota
    - Preset quota for defined category then recruit subjects
    - Bias: subjects whom are eligible but not recruited are different
  3. Snowball
    - Use selected participants’ friends, friends of friends
    - Bias: Others selected are similar, hence sample is not diverse (not generalisable to general ppn)
38
Q

Three main practical methods of questionnaire administration. Name some advantage/ disadvantages of each

A
  1. Mail/e-mail:
    - Low response (can increase by sending reminders, incentives etc.)
    - Not suitable for knowledge questions (can google)
    - Convenient, cheapest, most efficient
  2. Telephone interviewing method
    - Allow OEQ, complex interviews and is efficeint
    - Good response rate (65-75%)
  3. In-person interviewing
    - Allow OEQ, complex interviews with VAS (visual-aids)
    - Very good response rate (70-80%)
    - Time consuming
39
Q

What is turnaround time for data collection for questionnaire administration

A

Time taken for data to be collected, applicable for telephone and in-person interview

= (ns x t)/(ni x ti)

  • ns: number of surveys to be collected
  • t: duration for each interview
  • ni: number of interviewers
  • ti: No. of hours each interviewer work per day
40
Q

Common sources of secondary data that can be used for resh purpose

A
  1. Medical/Rx/Dispensing records
  2. Registry data
  3. Others: Databases/claims, XSS, large pCoh Data, spontaneous and surveillance data
41
Q

Describe the strength and limitations of using Medical/Rx/Dispensing records as secondary data

A

Good:

  • Efficient prospective data, less bias
  • Detailed info
  • Can study many drugs relating to many O
  • Can do nCCS
  • Has clinical AND lab data

Bad:

  • Unsure compliance to drugs (drug use)
  • No info on non-Rx drugs
  • Dx may be different ICD coding and may be unknown
  • No info on habits, hx and CF
  • Unknown data completion status
42
Q

State some examples of registry data. Describe the strength and limitations of using Registries as secondary data.

A

Examples: Disease registry, drug use reg

Good:

  • Cheap, if onee xists
  • Representative of ppn
  • Detailed and systematically collected

Bad:

  • Unsure drug use
  • Expensive, if creating one
  • May contain missing data
  • No appropriate control group from registry, require to source for other drugs or disease
43
Q

List and describe the considerations in secondary data collection and use

A
  1. Reliability: 5W1H on how data collected
  2. Suitability: Can answer YOUR resh qn?
  3. Adequecy: how complete data is? Got CF? Accurate enough?
  4. Ethical: PDPA? Need approval? Need de-identification?

Principles of Research (9 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions