exam

• systematic research is variables identified and research is designed to test the relationships between variables
• empirical research is collecting data on which to base decisions
• reductive research is using data to establish general relationships
• 5 types of research is ADEQM
• analytical research is in-depth study and evaluation of info to explain complex phenomena
• descriptive research is info collected without manipulation
• experimental research involves manipulation to establish cause and effect
• mixed methods is both qualitative and quantitative
• FINER is the criteria for a good research question
• Feasiable - adequate number of participants

  Intresting

  Novel - CONFIRMS or extends previous findings

  Ethical

  Relevant
• PICOT is the criteria for a good research question
• simple hypotheses predicts relationship between a single dependent variable and single independent variable
• complex hypotheses predicts the relationship between multiple dependent and independent variables
• directional hypothesis identifies the direction of relationship between variables e.g. positive
• associative hypothesis is change in one variable is associated with change in another
• casual hypothesis is manipulating the independent variable will change the dependent variable
• observational studies do not intervene
• experimental studies intervene, allows for cause and effect
• cross sectional assesses a phenomenon at one point in time, measures different populations
• cross sectional design is good to compare different groups of people
• case control study is usually retrospective. cases have the outcome, control does not have the outcome
• randomised controlled trial is prospective, participants are randomised into a control or intervention group, follows group over time to determine outcomes
• randomised controlled trials try to prove cause and effect
• quasi experimental design does not randomise participants or have a control group
• pilot studies are a small case study conducted prior to a large scale experiment to test and refine procedures
• top of hierachy of evidence is randomised controlled trials, bottoms is editorals and mechanistic
• stratified random sampling is the population is divided on a characteristic
• convenience sampling is researchers being interested in a specific type of person, this leads to snowball sampling
• mechanistic study is intervention study comparing two groups
• meta analyses converts data from studies into a standard metric
• small/medium scale RCT are to inform future large scale RCT
• systematic review assesses methods of other studies
• confounding bias is error or inaccuracy in the effect of an exposure on an outcome due to the influence of another factor
• residual confounding bias is error or inaccuracy that remains after controlling for confounding in design of a studfy
• conflict of interest is a situation in which a person is in a position to derive personal benefit and unfair advantage from actions made in their official capacity