ap statistics 樣本分類

1、AP Statistics Lu Hao2.3An experiment is a planned intervention undertaken to observe the effects of one or more explanatory variables. Explanatory variable(factor)Response variableExperimental condition/treatment: (combination of values for explanatory variable)Design of a experiment: overall planEx

2、traneous factorDirectly control(holding some extraneous factors constant)Blocking(create groups/blocks that are similar, try all treatment in each block)Randomization(random assignment)Replication(making multiple observation for each experimental condition)Definition of BlockingUsing extraneous fact

3、ors to create groups that are similar. All experimental conditons are then tried in each block.Example for BlockingAn experiment is designed to test a new drug on patients. There are two levels of the treatment,drug, andplacebo, administered tomale andfemale patients in a double blind trial. The sex

4、 of the patient is ablocking factor accounting for treatment variability betweenmales andfemales. This reduces sources of variability and thus leads to greater precision.The ideal situation would be to have both random selection of subjects and random assignment of subjects to experimental conditons

5、Homework P40, 19a,b,c,d P49, 32P67 62Read section 2.5 and 2.63.1 Comparative Bar ChartFor categorical dataUse relative frequencyPie chartCategorical dataSmall number of possible categoriesFor illustrating proportions segmented bar chart(P81)3.2 Stem-and-leaf displaysFor univariate numerical dataDoes

6、 not work well for very large data setsDef: Outlier an unusually small or large data valueStem-and-leaf displaysWhat to look for: gaps, symmetry, peaks, outliers representative or typical valueRepeated stems to stretch a displayComparative stem-and-leaf display(92)3.3Frequency distribution and histo

7、gram for numerical data Discrete variableContinuous variableHistogram for discrete numerical dataWorks well even for large data setExample 3.11(P97)Example 3.13(P99)Histogram for continuous variableHistogram works well, even for large data setsWhat to look for: central or typical value extent of spr

8、ead or variation general shape location and number of peaksPresence of gaps and ourliersCumulative relative frequencyCumulative relative frequency plotA cumulative relative frequency plot is just a graph of the cumulative relative frequencies against the upper endpoint of the corresponding interval.

9、We can know: what proportion of the observation is smaller than a particular value; what value separates the smallest p percent from the large values.Read P107- P108Histogram shapeUnimodal: single peakBimodal: two peaksMultimodal: more than two peaksSymmetric unimodal histogramA unimodal histogram is symmetric symmetric if there is a vertical line of symmetry such that the part of the histogram to the left of the line is a mirror image of the part to the right. Lower tail Upper tailSkewed unimodal hi