Statistics. txt

  1. Data
    Observations (such a measurements, genders, respones to a survey) that have been collected by various methods.
  2. Population
    is the complete collection of all elements (scores, measurements, responses, etc.) to be studied. The collection is complete in the sense that it includes all subjects to be studied.
  3. Statistics
    A collection of methods for planning studies and experiments, obtaining data, and then organizing, summarizing, presenting, analyzing, interpreting, and drawing conclusions based on the data.
  4. Parameter
    a numerical measurement that decribes some characteristic of a population.
  5. Statistic
    A numerical measurement that decribes some characteristic of a sample.
  6. Quantitative Data
    Consists of numbers that represent counts or measurements.
  7. Qualitative Data
    Consists of data that can be organized into different categories or groups according to some nonnumerical characteristic.
  8. Discrete Data
    Consists of quantitative data values that have only a finit number of values or infinitely many values that are countable.
  9. Continuous Data
    Consists of quantitative data values that can be any real number in some finite or infinite interval on the real numer line.
  10. Nominal Level of Measurement
    Data that consists of names, labels or categories only. There is no natural way to put data in some type of order.
  11. Ordinal Level of Measurement
    Characterized by data that can be arranged into some type of order, but the difference between two data value can not be determined or is meaningless.
  12. Interval Level of Measurement
    Very much like the ordinal level of measurement, with teh additional proprty that the difference between data is meaningful. However, data at the ordinal level has no natural data value equal to zero that represents a natural starting data value.
  13. Ratio Level of Measurement
    Is like the interval level of meaurement, with the additional property that there is a data value equal to zero that represents a natural starting point. With the ration level of measurement, it makes sense to subtract and divide data values.
  14. Common Misues of Statistics
    • Bad sample /Inappropriate sampling techniques.
    • Voluntary response sample
    • Small samples
    • Misleading graphs
    • Pictographs that distory reality
    • Loaded questions
  15. Common Misues of Statistics (Cont'd)
    • Wording of questions
    • Order of questions
    • Self-interet study
    • Correlation and Casualty
    • Refusals
    • Deliberate distortions
    • Precise numbers
  16. Observational Study
    Data is collected y observation and measurement of specific characteristics. There is no attempt to modify the subjects being studied.
  17. Experimental Study
    Some treatment is applied to the subjects and its effects on the subjects are observed.
  18. Confounding
    Occurs in an experiment when it is not posible to distinguish among the effects of different factors.
  19. Random Sample
    A random sample of a population is constructed so that every member of the population has an equal chance of being included in the sample.
  20. Simple Random Sample
    A simple random sample of n subjects is constructed in such a way that every member of the population has an equal chance of being included in every possible sample size of n
  21. Systematic Sampling
    A starting point is first selected and the every nth element is selected.
  22. Probabilty Sampling
    Involves selecting members from a population in such a way that each member has a known chance (not necessarily the same) of being included in the sample.
  23. Convenience ampling
    We simply use results that are easy to obtain, such as surveying your friends or having a person answer a question by clicking a button on an internet browser.
  24. Stratified Sampling
    Subdivide the population into two or more subgroups so that all members of a subgroup share the same characteristic, then draw a sample from each subgroup.
  25. Cluster Sampling
    Subdivide the population into non overlapping sections (or clusters) and randomly select some of the clusters. The sample includes all member of the selected clusters.
  26. Sampling Error
    Is the difference between a sample result and the true population result. Sampling errors are caused by chance alone because chance determines the random sample obtained. Sampling errors can be minimized by using larger random samples but can never be truly eliminated.
  27. Nonsampling Errors
    Caused by human error, such as inappropriate ampling methods, incorrectly collected data, improperly recorded or copied data, numerical miscalculations, or faulty data analysis.
Card Set
Statistics. txt
Chapter 1 terminology