| |
|
 Main
Concepts: Exploring Data
• In this course we'll see two approaches to data analysis. In
one, we ask questions and then gather data to answer the questions. In
the other, we have data already, and then ask appropriate questions. In
this unit we'll focus on the second approach.
• Data are numbers in context.
• In order to answer questions about research or study, data must first
be organized.
• Organizational tools used depend on the question and the nature of
the data. For example, boxplots are very useful for comparing two or
more groups. Dotplots are useful for viewing the distributions of small
data sets, and histograms are useful for viewing the distributions of
larger datasets. But keep in mind that these are very general
guidelines, and experimentation is always encouraged.
• Numerical and graphical summaries should capture features of the
distribution of the data. Important features include the center, the
spread, and the shape.
• Summaries (both numerical and graphical) have dual roles: they help
us explore and discover possibly unknown features of the data, and they
are also used to communicate features of the data to
others. .
• Focus should be on interpretation, not calculation (of numerical
summaries) or construction (of graphs).
• Graphs do not speak for themselves. Students (and you)
should never provide a graph without an interpretation.
|
