stats terms

• Parameter - characteristic of a population which is usually unknown and needs to be estimated
• statistic - computed from a random sample and used to estimate unknown parameter
• estimator - formula used in coming up with an estimate
• estimate - numerical value arrived when applying estimator using sample data
• point estimate - specific numerical value of a population parameter
• interval estimate - range of values that may contain the parameter of a population
• accuracy - measure of how close the estimates are to the actual value of the parameter
• precision - measure of how close the estimates are with each other
• sample - construct estimates for the parameter of the population
• sampling distribution - bridge that takes us from the probability to statistical inference
• statistics - measure that describes a sample
• sampling - process of selecting units from a population
• cencus - process of gathering information from every unit in the population
• advantages of sampling:

  • reduced cost
  • greater speed
  • greater scope
  • greater accuracy
• simple random sampling - method of selecting n units out of N units in the population in such a way that every distinct sample size n has an equal chance of being drawn
• stratified random sampling - population of N units is divided into subpopulations called strata
• equal allocation - the sample sizes from the different strata are equal
• proportional allocation - the sample sizes from the different strata are proportional
• systematic sampling - random start; method of selecting a sample by taking every kth unit from an ordered population
• k - sampling interval
• 1/k - sampling fraction
• cluster sampling - sampling method in which the population is divided into clusters
• purposive sampling - make a sample agree with the profile of the population based on some pre-selected characteristics
• quota sampling - selects a specified number (quota) of sampling units possessing certain characteristics.
• accidental/haphazard/convenience sampling - selects sampling units that come to hard or are convenient to get information from
• expert or judgement sampling - selects sample in accordance with an expert’s judgement
• modal instance sampling - selct samples from specific groups or types of people wherein sampling the most frequent case is sought for
• diversity or heterogeneity sampling - opinions/views of a specific topic are the primary concerns and representing the views proportionately is not of major importance
• snowball sampling - selects a sample who meets the specified criteria; respondents are asked to recommend someone
• estimation - process of determining parameter values
• bias - the difference of the expected value of the estimate from the true value of the parameter
• critical value - value is compared to determine whether or not the null hypothesis is rejected
• null hypothesis- no difference
• significance level - degree of certainty required in order to reject null hypoin favor of the alternative hypothesis
• two-tailed test - both sides
• type 1 error - true and rejected
• type 2 error - false and accepted
• alternative hypothesis - difference exists
• one-tailed test - one side
• correlation - determine relationship between variables