100%(4)4 out of 4 people found this document helpful
This preview shows page 7 - 10 out of 94 pages.
Bars show frequency of scores in each 5% range■More close together / higher grades = best class■More spread apart = more variety in scores●Constructing a Frequency Distribution Using Grouped Qualitative Data○What it means■Numbers that aren’t whole numbers à need to group these number sets into classes to limit results○Number of Classes■2k > nrule-k = number of classes-n = number of data points■Find lowest value of k that satisfies that rule■Ex: 50 data points-2k > 50-2^5 = 32 > 50 (too small)-2^6 = 64 > 50 (perfect)●Class Width○Determining Class Width■Width = breadth (range) of numbers we plan to put into each classEstimated class width = (max data value – min data value) ¸ k○Example: where k equals 6, max data value equals 17.4, and min data value equals 0.6■(17.4 – 0.6) ¸ (6) = 2.8○Other Rules
■Round to whole number■No correct answer (can round 2.8 up to 3 or 4)-b/c trying to identify useful pattern in the data, and there’s often more than one way to do this●Class Boundaries○Class Boundaries– the minimum and maximum values for each class○Choosing Class Boundaries■Try to stick with whole numbers■Stick to your class width○Ex: min data value equals 0.6 and class width is 3, and number of classes equals6■0 to less than 3■3 to less than 6■6 to less than 9■9 to less than 12■12 to less than 15■15 to less than 18○Ex: min data value of 0.6, class width of 4, and number of classes of 6-Allows 6 classes, but okay if less than 6 classes-Below = if we’d chosen class width of 4, rather than 3■0 to less than 4■4 to less than 8■8 to less than 12■12 to less than 16■16 to less than 20●Class Frequencies○How to find it■Count number of observations in each class and record the total○Rules: Constructing Classes for Grouped Data■Equal-Size Classes-all classes in the frequency distribution must be equal width■Mutually Exclusive Classes-The class boundaries cannot overlap, b/c specific data points can end up in more than one class-Ex: 0 to less than 3 / 2.5 to less than 5.5 (NO)■Include All Data Values■Avoid Empty Classes-Don’t display classes so narrow that there are no observations in it
■Avoid Open-Ended Classes (if possible)-Example of open-ended class: 18 and over (rather than 18 to less than 21)●Constructing a Histogram with Grouped Quantitative Data○Steps for Creating Histogram■Use continuous data listed in table 2.6 (page 30)■Same steps as listed before for Excel (data used listed below)○Edits for Histogram of Continuous Grouped Data■Format the data series and drag the “gap width” to the left■Format Axis > Alignment > Custom Angle (-45)●The Consequences of Too Few or Too Many Classes○To Few Classes■Happens if use to wide of classes■à obscures patterns■Shows to little data (ex: if class width of 9, may only have two classes) à only two bars in your histogram○To Many Classes■Happens if too narrow of classes■