Everything Flashcards by Carys Hooper

What is a p value?

Calculated probability of observing an effect size as large or larger than the one observed between the groups, if there was no actual difference between the groups

How well did you know this?

Not at all

Perfectly

How is qualitative research carried out?

Interviews
Participant observations
Focus groups
Textual analysis
Action research

How well did you know this?

Not at all

Perfectly

What are the key terms in qualitative research?

Interpretation
Reality
Truth
Generalisation

How well did you know this?

Not at all

Perfectly

What is quantitative research?

Numbers,
Testing a theory
Objectivity
Large sample size
Generalisable narratives

How well did you know this?

Not at all

Perfectly

What does quantitative research give you?

Clear answers to specific causal questions

How well did you know this?

Not at all

Perfectly

What is qualitative research?

Meanings 
Social life
Generating theory
Subjectivity
Small sample size

How well did you know this?

Not at all

Perfectly

what does qualitative research leave you with?

Applicability within context

How well did you know this?

Not at all

Perfectly

What are the cons of qualitative research?

Researcher presence and bias
Time consuming
Lack of generalisability
Poor replication and validity

How well did you know this?

Not at all

Perfectly

What happens in an observational study?

No intervention

Record behaviours, attitudes and symptoms

How well did you know this?

Not at all

Perfectly

What is a cohort study?

Group of subjects followed over time

Done to find the incidence and investigating potential causes of condition

How well did you know this?

Not at all

Perfectly

Are cohort studies prospective or retrospective?

Either

How well did you know this?

Not at all

Perfectly

What is an experimental study?

Researcher controls or introduces factor

Record effects of intervention on outcomes

How well did you know this?

Not at all

Perfectly

What is a parallel study?

Treatment vs placebo

How well did you know this?

Not at all

Perfectly

What is a crossover study?

Use the same two groups and do a period when they are the placebo (and then a wash out period) and then a period when they are the active treatment group

How well did you know this?

Not at all

Perfectly

What are the phases of a preclinical trial?

Preclinical
Clinical (phase 1-3)
Post clinical

How well did you know this?

Not at all

Perfectly

What happens in the preclinical phase?

Lab tests
Research protocol
Non-human testing

How well did you know this?

Not at all

Perfectly

What happens in a phase 1 trial?

Small group (20-100 healthy patients) 
Dose finding

How well did you know this?

Not at all

Perfectly

What happens in a phase 2 trial?

Large group (100-1000s) 
Testing of beneficial and undesirable effects

How well did you know this?

Not at all

Perfectly

What happens in a phase 3 trial?

1000+
Compare with standard therapy
Multiple worldwide locations

How well did you know this?

Not at all

Perfectly

What happens in a post clinical trial?

Analyse data and write a paper
Follow up
New therapy shown to be safe and effective, approved by medical authorities and continuing testing

How well did you know this?

Not at all

Perfectly

What are the different parties in a clinical trial?

Patients
Clinicians
Institutions where trials are held
Research ethics committees or the institutional ethical committee 
Sponsor
Regulatory authorities

How well did you know this?

Not at all

Perfectly

What is the step by step of running a clinical trial?

Recruitment
Screening
Informed consent after patient information sheet
Blinding
Protocol visits step by step procedures
Reports of adverse effects 
Store investigates product 
Subject compensation
Electronic data collection 
Study closure

How well did you know this?

Not at all

Perfectly

What are the two main cases that have influenced human research protections?

Public Health Service syphilis study

Nuremberg (nazi doctors/ww2 scientists)

How well did you know this?

Not at all

Perfectly

What does the Nuremberg code state?

Need for informed consent
Research should be based on prior animal work
Risks should be justified by anticipated banality
Only qualified scientists must conduct research
Physical and mental suffering should be avoided
No research where death/severe injury is expected

How well did you know this?

Not at all

Perfectly

What are the two methods of sampling?

Simple random | Convenience

What are the pros/cons of simple random sampling?

Every member has an equal probability of selection Representative sample Not particularly feasible or cost effective though

What are the pros/cons of convenience sampling?

Participants convenient to access | Sample not particularly representative and you may get selection bias or volunteer bias

What are the different types of variables?

Categorical | Continuous

What is categorical data?

Variables with options that don’t have a predictable quantitative relationship between them

What is continuous data?

Variables measured on a quantitative scale

What are the types of bias?

Response bias | Assessor bias

What is response bias?

Difference between response and truth | Particular issue if you are self reporting issues

What is assessor bias?

Systematic difference between assessor measurement of participant and truth

What is the 95% central reference range for numerical data with normal distribution?

A measure of the spread of the continuous numerical data

What is the unpaired t test used for?

Analysing parametric data from independent samples

What is parametric data?

Data that you’ve assumed normal bias for

What is the paired t test used for?

Paired parametric data

What does the one tailed test do?

Only care about significance in one direction and will talk about significantly greater then or significantly less than

What does the two tailed test do?

Care about significance in both directions and talk about significant difference

What does ANOVA stand for?

ANalysis Of VAriance

What is ANOVA?

Standard approach for statistical analysis of studies involving multiple comparison

What does ANOVA assume?

Data is parametric and unpaired

What is a one way ANOVA used for?

Studies with 3 or more varying conditions on a single continuous variable

What is a two way ANOVA used for?

Studies where there is an interaction between two categorical variables on a single categorical variable

When are post tests used?

If ANVOA is significant

What are the types of categorical data analysis?

McNemars test Fishers exact test Pearsons chi-squared test

What is McNemars test appropriate for?

Analysing continuous data involving paired data that can be shown in a 2x2 contingency table

What is a fishers exact test?

Analyses categorical data involving unpaired data that conform to a 2x2 contingency table and a small number of samples

What does Pearsons chi-squared test analyse?

Categorical data involving unpaired data that cant be measured with fishers test

When cant you use Pearsons chi-squared test?

If any cell contains zero

What is a population?

Entire group of people

What is a sample?

Little group of the population

How can you work out they 95% confidence interval for a single mean?

Mean +/- 2 standard errors

What is the 95% confidence interval a measure of?

Precision

How do you work out relative risk?

Risk in group 1/risk in group 2

How do you work out risk difference?

Risk in group 1 - risk in group 2

What is the four principles approach in ethics?

Non-male Vicente Beneficence Autonomy Justice

What is civil law?

Dealings between private individuals or groups Rights and duties owed by individuals and groups to each other Legal action taken by claimants

What is criminal law?

Matters serious enough to be considered offences against the whole community Rights, duties and norms important for the whole community Legal action taken by the crown prosecution service

What is statute law?

Written law decided by legislature or another government agency Relatively difficult to change