Скачать или смотреть Basic course in biomedical research-Assignment 10

Data Collection tools
1. Review of records  Personal records, Clinical records, surveillance records, registers
2. Structured observation guide  Checklist of items (Text, figures)
3. Questionnaire  Interviewer admin. (face to face, telephone, Computer assisted); Self admin. (Paper-based, computer-assisted)
A sociogram, on the other hand, is a diagrammatic tool used primarily in social network analysis to visualize relationships and social structures — it's not a method of data collection, but rather a way to represent data, usually from quantitative or mixed-methods studies.
Info. Collected with data collection tools
Facts  Indiv. Char, env, behavior practices; Knowledge  Risk factors, healthy lifestyle; Judgement  Opinions, attitude
Pilot study  NOT hypothesis testing. NO inferential statistical tests, NO p-values, Small samples. Observations from a pilot study can be added to the final sample size
Primary factors to be considered for calculating a sample size?
1. the effect size (usually the difference between 2 groups)
2. Population standard deviation (for continuous data)
3. Power of the experiment to detect the postulated effect
4. Significance level
Independent Variable (IVC) can be changed  The variable that is manipulated or controlled by the researcher to observe its effect on the dependent variable. Think of it as: The cause. Example: In a study on the effect of study time on test scores, the study time is the independent variable (because you control or change it).
Dependent Variable (DV)  Measured or observed to assess the effect of the independent variable. Think of it as: The effect or the outcome. Example: In the same study, the test score is the dependent variable (because it depends on how much time was spent studying).
Covariates  Potential risk factors or confounders
Tools to collect data (Google forms or Epicollect)
1. Abstraction form  Review of records (Personal, clinical, surveillance) Registers
2. Structured Observation Guide (Text, figures)
3. Questionnaire
a. Interviewer admin.  Face to face, Telephone, Computer assisted
b. Self-admin.  Paper based, computer assisted
Key elements of data collection tools (Clarity, balance, length, comprehensive [response], constraints [response], utility [instructions], order [questions], context [questions])
4 components of data collection tools
1. Intro. & conclusion
2. Identifiers (Name, address, ID number)
3. Instructions for data collectors (Prompts, skip patterns, use different fonts)
4. Body of the instrument (open, semi-open, closed)
Open Qs  Answers not suggested; respondent must generate answer (What d/s can you acquire from tobacco, what places did you eat at in the week preceding the disease ?); useful at hypothesis raising stage
Open Qs with closed end answer  No option of answer is suggested; Expressed as open Q, analyzed as closed end Q; eg. What are the practices that may increase your risk to get a heart attack ? (DO NOT propose any option of answer)
Closed Qs
1. Dichotomous Qs  Y/N
2. MCQs
Semi-open Qs  Suggested answers; MCQs + Other, specify : ______
Qualitative close ended questions have categorical responses (eg. Yes/no, high/medium/low, red/blue/green/pink, etc). These are analysed as frequencies/proportions /percentages in descriptive analysis and logistic regression for risk factor analysis.
Avoiding negatives and double negatives ; use neutral tone ; ambiguity  the quality of being open to more than one interpretation; inexactness. Go from casual to intimate
Skip patterns (also called routing or branching) are important in questionnaires to direct respondents through relevant questions based on their previous answers.
1. Open-ended question: No restrictions on response length or type.
2. Closed question: Provides fixed answer options.
3. Semi-open question: Allows respondents to answer freely but with some restrictions (e.g., number of responses).
4. Open question with closed answers: Usually contradictory; open questions don’t have closed answers.
Reliability  Repeatibility/precision. If repeated, gets same results; Accuracy  Validity