Epidemiologist Skills PDF

Skils Neededfor Epidemiologist(Perior caring on the study) Title should reflect exactly and simply Study Problem, Questions , Design and type.include Place and time Skils Neededfor Epidemiologist(Perior caring on the study) Deep Knowledge in Epidemiology Good Problem...

Skils Neededfor Epidemiologist(Perior caring on the study) Title should reflect exactly and simply Study Problem, Questions , Design and type.include Place and time Skils Neededfor Epidemiologist(Perior caring on the study) Deep Knowledge in Epidemiology Good Problem Defination and clear Questions=Desigen Methodology andTest Releations Good knoweldge in Stastical hypothesis forrmoulation(Testing hypothesis),and how usingStastics in Researchs, and Stastical infrences Research Design Tips Clear hypothesis formulation(Title)need clearknowledge in Problem defination and Questions. Appropriate sample size(Reprsentative) Proper test selection Power consideration Error prevention strategies Awaring regarding Sampling and Statistical Symbols N = Total population size,n = Sample size,p = Probability,μ = Population mean,σ = Population standard deviation,x̄= Sample mean Population vs Sample Population: Complete set of subjects (N) Sample: Subset selected for study (n) Example: City population N = 500,000 Study sample n = 1,000 Descriptive Statistics, you should remember(Numbers) in Quantitife studies Measures of Central Tendency: Mean (average) Median (middle value) Mode (most frequent) Measures of Variability(SampleTrind) Range Variance Standard Deviation Used to understand data spread Inferential Statistics are about Making predictions about populations using: Hypothesis testing(rejecting NO When PV α: Fail to reject H₀ Example: p = 0.03 < α = 0.05 Avoid Type I Error(regarding Results and reality) False Positive Rejecting H₀ when true α = probability of Type I error Example: Concluding treatment works(Positive) when it doesn't(False) Avoid Type II Error(regarding Results) False Negative Not rejecting H₀ when false β = probability of Type II error Example: Missing real treatment effect(Negative false) Statistical Power to Avoid Systematic Errors(in The Sample=Beta)=Ensure representitive sample Power = 1 - β Factors affecting power: Sample size Effect size Significance level Chose adequate test: Z-tests Use when: n > 30 Population σ known Tests population mean differences Chose adequate test: T-tests Overview Use when: n < 30 Population σ unknown Three types available Types of T-tests One-sample Independent samples Paired samples Each serves different purposes for Precision: Confidence Intervals Part 1 Range containing true population parameter Usually 95% confidence level Interpretation basics Confidence Intervals Part 2 Factors affecting width: Sample size Variability Confidence level Sample Size Considerations Larger samples: Increase precision Reduce error Improve power Common Statistical Tests When to use: Z-test vs t-test Paired vs unpaired Parametric vs non-parametric