RESEARCH

• Descriptive statistics


  • Used when you want to know something about everyone in an entire group
  • Describes or summarizes the data of a target population
  • Describe data that is already known
  • Organize, analyze and present
  • Results are shown in tables and graphs
• Inferential statistics


  • Used when you want to know something about everyone from a smaller group
  • Making a generalization about a larger group of individuals on the basis of a subset
  • Use data to make inferences or generalizations about the population
  • Make conclusions for population that is beyond available data
  • Compare, test, and predicts outcomes
  • Final results is the probability scores
• Tools for descriptive statistics

  • Measures of central tendency
  • Measures of dispersion
• Tools for inferential statistics

  • Hypothesis test
• Parameter Estimates - seeks an approximate calculation about a feature of a population 

  “By how much does this new drug delay relapse?”

  • Point estimate - one number (the point), the same for the population as for sample

  Interval estimate - range of numbers, including the point, an
• Hypothesis Testing - seeks to validate a supposition based on limited evidence, inferred using a sample from the population 

  “Does this new drug delay relapse?”

  • Null hypothesis - what is happening is due to chance (no relationship)
  • Research/Alternative Hypothesis - not chance, something is going on 
• HYPOTHESIS TESTING 

  • To find out whether the observed variation among sampling is explained by sampling variations, chance or there is  really a difference between groups

  Significance test - the method of assessing the hypotheses testing whether a result is likely to be due to chance or to a real effect
• Steps

  1. State hypothesis
  2. Select level of significance 
  3. Identify test statistic
  4. Formulate decision rule
  5. Take a sample, and decision 
  6. Do not reject Ho
  7. Reject Ho 
  8. Accept H1
• Null hypothesis - Ho : x1 = x2 this means that there is no difference between x1 & x2 

  • The hypothesis that researcher tries to disprove or nullify 
  • If we reject Ho, it’s either H1 : x1<x2 or H2:x1>x2

  Alternative hypothesis

  • The hypothesis that the researcher tries to prove
• type I error (false-positive) - occurs if an investigator rejects a null hypothesis that is actually true in the population;

  type II error (false-negative) - occurs if the investigator fails to reject a null hypothesis that is actually false in the population.