Causality Flashcards
Pros of samples
can be studied more quickly than larger populations
Hypothesis testing
based upon what is commonly believed (the status quo)
Alternative hypothesis (H1)
statement that contradicts the null hypothesis
6 steps for evaluating hypotheses
- Formulate the Ho in statistical terms (Ho: Odds ratio = 1)
- Formulate the H1 in statistical terms (H1: Odds Ratio = 1).
- Select the level of significance for the statistical test and sample size (generally 0.05 or conservatively 0.01)
- Select the appropriate test stastic and identify the degrees of freedom and the critical value
- Collect the data and estimate the measur eof association and test statistic
- If the observed measure exceeds the critical value, reject Ho in favor of H1; otherwise do not reject Ho
The null hypothesis is assumed to be correct unless
there is sufficient evidence from the sample data to indicate otherwise
A tentative suggestion that a cerain phenomenon exists
research hypothesis
P value
the probability for evaluating the role of chance; ranges from 0 to 1
Small p value (> or = 0.05)
the result is unlikely to be a product of chance
Confidence intervals
range of reasonable values in which a population parameter lies that is based on a random sample from the population
When sample size increases
the role of chance decreases
To minimize chance
increase sample size
Bias
deviation from the truth
Random error
incorrect result due to chance
Systematic error
incorret result due to bias
Confounding
a third factor that influences the relationship between an expoure and disease outcome
Determinants
causes/factors
Epidemiology
the study of the distribution and determinants of health-related states or events in human populations and the application of this study to the prevention and control of health problems
Identifying what determines (cause) disease allows us to
prevent and control the health problem rather than react to it