Lecture 1 and 2 Flashcards
(24 cards)
what is statistics?
statistics is the science of learning from data, and of measuring, controlling and communicating uncertainty
biostatistics?
Biostatistics is the development and application of statistical methods to answer research questions in the health sciences.
Who holds the responsibility for making sense of data?
researcher, journalist and consumer
The scientific method
The scientific method is a process we use to gain knowledge and understanding through testing of hypotheses.
Statistics provides much of the ….
framework and mechanisms for implementing the scientific method.
order of events in the scientific method
start with observation and then form a hypothesis (based on the observation) then make predictions and then design and carry out an experiment/study to test hypothesis (collecting data) and then lastly accept/reject/refine theory (interpret results)
The scientific method aims to get you closer to the scientific truth
The importance of reliable evidence
The reliability of the evidence regarding our hypothesis is critically dependent upon the quality of the research.
- Evidence based policy
- Evidence based practice
Evidence based policy
Aims to apply the best available evidence gained from the scientific method to policy making.
evidence based notion used in policy e.g. government such as the bowel screening programme
quality of diagnostic test (efficacy) checked before policy changed is an example of this
Evidence based practice
Preferential use of interventions or actions for which systematic empirical research has provided reliable evidence of benefit.
în practice means at the individuals level for example in a clinical trial
Health sciences research goal
Goal: to improve the health and well-being of individuals and communities
Health science research involves
This involves…
prevention of disease
treatment of disease
promotion of health
In order to do health science research we need knowledge about…
causes of disease diagnosis of disease disease process (natural history) effective treatments societal factors which affect health
wrong/inaccurate data =
wrong/misleading conclusion
Research is a process for
providing reliable answers to questions for which the answer is not immediately available.
Scientific research provides a
systematic process for obtaining reliable answers to such questions
reliable = reproducible research
defining the research questions leads to quantifiable evidence
five steps of the research process
1 Development of research question (ends to be well defined) 2 Design of study 3 Collection of data 4 Description/analysis of data 5 Interpretation of results
all of these steps are essential and none can be skipped
specification of the research question
The specification of the research question will include determination
of a model for the process under study, and any hypotheses to be tested.
what does the study design refer to?
The study design refers to the methods used to select the study participants, control any experimental conditions, and collect the data.
how we collect data, from which group
Inference
The aim of most research studies is to use data from a sample to draw inference about a particular characteristic of a larger population.
cycle is underlying population - sample - statistics - inference - underlying population etc
How do you get the best guess at true data?
Best guess at the true data is done by using your sample
you do not expect to be as accurate as if you tested whole population
with successive samples estimates would express variability
Model
A model describes the relationships between the quantities of interest in a population
Variability and uncertainty
There is variability (or natural variation) in almost everything we may wish to measure
This leads to uncertainty about true value of the quantity or relationship we wish to measure.
Statistics provides a framework for answering questions from data in an atmosphere of uncertainty
We describe this uncertainty using the language of probability
A statistical model takes account of the uncertainty, and includes a mathematical description of the way the data are generated.
what ever we are measuring is subject to random variability
A statistical model takes …
account of the uncertainty, and includes a mathematical description of the way the data are generated.
describes the relationship between multiple variables of interest
bias as an area is easy or difficult to deal with?
it is difficult to deal with this systematic error and it affects the results
