Skip to content | Accessibility - list of access keys used on this site | Sitemap
Find what you need to know

Home > Subjects > Mathematics > Level 3 - statistics and modelling > 3.5 Statistical investigation (AS90645) > Achievement criteria

  • Subject: Mathematics
  • AS: AS90645
  • Level: 3
  • Credits: 3
  • Internal

Statistics and modelling 3.5 Select and analyse continuous bi-variate data

Achievement criteria

The level of achievement that you reach is decided by the questions that you answer correctly.

Achievement

  • You select and analyse continuous bi-variate data.
  • You may collect the data or it may be provided. It should be data for which a linear model is appropriate. Where the data is provided it will involve more than one pair of variables from which you select a pair.
  • The analysis will involve:
    • developing a purpose statement from the data selected
    • graphing data
    • using regression to establish a linear relationship between a pair of variables
    • describing the relationship between at least one pair of variables in context.

Achievement with Merit

  • You need to reach Achievement.
  • You carry out an in-depth analysis of bi-variate data.
  • Your analysis will include some of the following:
    • comparing the relationship between more than one pair of variables
    • discussing the appropriateness of the model
    • interpreting correlation coefficients, r, and coefficients of determination, R2, when appropriate
    • making predictions from regression equations (interpolation and/or extrapolation)
    • use of residuals
    • discussing the difference between correlation and causality when appropriate.

Achievement with Excellence

  • You need to reach Achievement with Merit.
  • You report on the validity of the analysis.
  • The report will include justified comments on some of the following:
    • method(s) of analysis
    • assumptions made
    • limitations
    • improving regression models, for example discussing the effect of outliers, fitting piecewise or non-linear models
    • alternative approaches
    • data source or data collection method if you collect your own data
    • potential sources of bias
    • relevance and usefulness of evidence
    • how widely the findings can be applied.

Back to top

3

People logged in

52

People surfing this site