Lecture Notes: Randomized Control Trials and Sampling

Acknowledgement of Country

• Students are encouraged to read the acknowledgement of country.

Learning Outcomes

• Randomized Controlled Trials (RCTs): understanding what they are and how they are designed.
• Sampling: understanding random versus non-probability sampling.
• Power Calculations: determining the necessary sample size.

Kahoot Quiz

• Used to assess understanding of pre-learning materials.
• Accessible via Kahoot.it and a game PIN.

Question 1: Characteristics of Probability Sampling

• Probability sampling is also known as random sampling.
• Qualitative approaches often use non-random sampling, especially when studying feelings, experiences, or cultural beliefs.
• Random sampling (probability) means every member of the population has a known chance of being selected (e.g., names in a hat).
• Non-random sampling (non-probability) is common in qualitative research because a sampling frame is often not available.

Question 2: Probability Sampling Approach

• Cluster sampling is a probability sampling method.
• Convenience sampling: selecting readily available participants. This is a non-probability method.
• Snowball sampling: participants refer other potential participants, often used with hard-to-reach populations. This is a non-probability method.
• Quota sampling: selecting participants based on predefined quotas (e.g., five females, five males). This is a non-probability method.
• Cluster Random Sampling: the full name. To do a cluster random sample, you need to have all the names in the hat and then to pick people randomly from that cluster.
  • Divides a map into sections (e.g., postcodes). Selects clusters randomly, then recruits all eligible participants within those clusters.

Question 3: Sampling - Snowball

• Snowball sampling involves identifying initial participants who then recommend others, especially useful for hard-to-reach populations.

Question 4: Aims of Quantitative Research

• Quantitative research aims to demonstrate cause and effect, relationship for prediction based on correlation.
• Qualitative research aims to understand individuals’ unique perspectives and how they define a phenomenon.
• Quantitative research describes the world beyond individual experiences.

Question 5: PICO

• PICO acronym: Population, Intervention, Comparison/Control, Outcome.
• P: Population of interest.
• I: Intervention being studied.
• C: Comparison or control group (non-intervention).
• O: Outcome being measured.
• Used to formulate research questions in quantitative studies (relationship between independent/intervention variable and dependent/outcome variable).

Question 6: Experimental Designs

• Experimental designs provide better evidence of causes and effects compared to non-experimental (observational) designs.
• Non-experimental designs involve observing participants without intervention.

Reading Journal Articles

• Goal: Become proficient in reading and appraising articles.

• Structure of a Journal Article:

  • Title

  • Abstract: Provides a summary of the article (background, objectives, design, participants, methods, results, conclusion).

  • Introduction/Background: Provides context, literature review, identifies gaps, and states the study's purpose and hypothesis.

  • Methods: Describes the study design (e.g., randomized control trial), sampling methods, and measurement tools.

  • Study Design examples:

    • Randomized control trial is study design
    • Cohort is study design
    • Case control is study design
    • Correlational study is study design

  • Methods is a very juicy section where you get a lot of information

  • in the sampling so in the methods, you know the sampling

    • Like, you can have a best guess of what the sampling is, but you look at the descriptions in this message sections to then tell, oh, I think the this is snowball or this is convened or this is cluster random sampling.

  • What do you use to measure sugar?

    • weight scale is known measurement tool or the outcome measure. Then, the word outcome. It's a measurable item. So, sugar is the outcome.

  • Consult Diagram: A flowchart showing participant recruitment, allocation to groups (intervention, control), and withdrawals.

  • Results: Presents statistical analyses (tables, p-values). In week four, week five, you will learn how to read a table like this and what does the P stands for.

  • Discussion: Summarizes key results, acknowledges limitations, and suggests future research directions.

  • Conclusion

  • References

Journal Article Activity: Alzheimer's Study Example

• Extract information from the abstract, background, and methods sections.

Identifying the Aim

• Extract the aim or objective of the study from the article.

Study Design Appropriateness

• Determine if a randomized controlled trial design is suitable for the study's objective.

PICO Elements

• Identify the Population (P), Intervention (I), Comparison (C), and Outcome (O) in the study.
  • Population of interest allows us to generalize the results to a wider population.

Research Question Formulation

• Formulate a research question based on the PICO elements. So often is there a relationship between this intervention and that outcome compared to this other group amongst the population of interest?
  • Research question = It has to sound like a questions and it's close ended.
  • close ended means you don't leave the questions getting people think. How, why no? It's is it?
  • Answer will be either yes or no.
  • Use statistical analysis to decide on solution of research question (yes or no).
  • When change to form of statement: hypothesis.
  • Hypothesis: belief, hoping this is true and will say YES to question.
  • The hypothesis: will be there is a relationship between the MBSR and depression, anxiety, stress, mindfulness compared to usual care amongst Korean nursing students.

Observational vs Experimental Studies

• Experimental studies (with interventions) are better for assessing the effectiveness of something because you have the participants to try, not just observing. Experiments, you can only and best see the effect.
• Experimental studies = involves and intervention. With the acronym, which acronym should we use? The P I C O. Yes. The pee pee.

App applicability

• Assess if the MDSR study in the article is applicable to Yana's situation. Factors to consider gender, study, discipline, age.

Table One: Provides Informations

• Table One: provides more information about the people in the sample.
  • m = mean
  • s d is a standard deviation.
  • \X^2 = Chi Square
• The numbers (P number) tell if the two groups are same or different, should be orange or orange and statistically have to be comparable.
  • If bigger than 0.005, then it's telling us that statistically these are the same.

Study Design and Sample

• Inclusion criteria: who should be involved in the study.
• Exclusion criteria: who should not be in the study.

Method Section: To determine this

• What does it say

How Many of These Students

• What was the inclusion and exclusion criteria for the sample?
  • Undergraduate nursing students first to fourth grade. Not engaging in any any of these. But an exclusion criteria will be someone that has exposed to previously an MBA something like that. So they weren't able to commit time it's more of a personal reason but not the characteristics of the person.

Probability Sampling and nonProbability Sampling

• Are you are you clear with this design? How how How old would you like me to go through or you happy to read it yourself? Sorry, Later, you go through? Okay, okay. So The main difference is which one you need all the names and the hats? Which which one do you actually need all the names in the hat The sampling frame.
  The probability sampling. When you have everyone's name in the hat, you can choose people randomly that's what that random means. But the nonprobability, you don't know who. So you're not able to calculate the probability of selecting this one person, what what is it? So you cannot. That's the main difference.

Sample SIze

• Sample Size: It means the statistical software does. They will tell them you put whatever you are looking for population wise and with the given information and then they will say just one look can check yes that power has that can help you in the end what results you are going to get. So so it's a very important number. Not not too low. When it's too low you will not get the results that you are looking for. So you are power calculations will justify how people you will need.

Process from Start with RCT

• Put a population size = who will have the inclusion exclusion criteria
• Follow and get samples and try to recruit close or as close to what you can
  • This will tell if is experimental/randomized.
• Look ahead when they see changes between both pre post measurement.
  • Some measurement can not prospective so you have to look ahead.

What is the Minimal Number of the Participants?

• That says per power of calculated and said would want 26 people within participants to detect for different outcome.

Percentage Drop-Out in Both Measurements.

• That will be the calculated for answer.

Assessment One Overview

• Part A:
  • Design an infographic (one page poster) containing key study details PICO. Inclusion, exclusion, power, sample and measurements.
• Part B:
  • Complete a critical appraisal report form (web document attachment).
    • The 12 questions for the whole part, but just want the question 10 for the week now.

How and What To do for Assessment:

• Design software, such has Canva, or the PowerPoint will give the information.
  • If needing in PDF, convert to make all into one PDF for submission.
  • So it does not send to multi Submission.
    1 submission is all.
• Due: week 7.