Weekend+One

Weekend One :: Saturday

 * What are the implications for students, teaching and policy?
 * What is statistics? A set of analytical tools to used to take a look at relationships and questions about data.


 * Types of Data**
 * Nominal - Classification or Categories - #'s have no inherit mathematical meaning.
 * Ordinal - Order (SA, A, D, SD), Class Ranking, GPA, 1st Vs. 3rd, Etc. Does not reflect the actual distance between ranks.
 * Interval Scale - Equal intervals between increments. Meter stick, grading scale, time
 * Ratio Scale - Same as interval but with a true zero. Example is temperature scale (degrees F or C, which have a 0 degree mark).


 * Frequency Distribution** - Describes how often an occurrence takes place. Can but used with all four types of data.
 * Histogram** - "Nothing more than a bar graph."


 * Class Notes**
 * All inferences or conclusions in stats needs to be supported with the data. Percentages can be used because they are a common ground when comparing.
 * Also relate back to policy, real-world, or setting. Great for the conclusion paragraph.


 * Assessment #1** - Grade distribution explained
 * Be sure to include the sample size - how many kids were sampled?
 * Relate back to the policy, real-world, or the setting.


 * "Problematic Data"**
 * "Skewed data" (see graphs on page 44) - The reliability of skewed data depends o:
 * Nature of the data
 * Sampled Population - Skewed is reliable for a given population, but is not reliable for a sample of the general population.
 * If something is negatively skewed, then MOST of the occurrences happen out to the postive end of the x-axis. The rare occurrences happen closer to the origin.
 * If something is positively skewed, then MOST of the occurrences happen closer to the origin and the rare occurrences happen out to the positive end of the x-axis.


 * Purpose of Statistics**
 * 1) Descriptive
 * 2) Inferential


 * Weekend One :: Sunday**
 * Mean - Average, Skewed data will thrown off the mean (outliers). Floor, Ceilin.
 * Once the Data is skewed, the mean is reliable.
 * Median - Middle Value
 * Mode - Most. Can be bimodal, but nobody uses mode.


 * Variance**
 * Statistics describes the variance, analyzes the variance.
 * The differences observed in the data.

What is the overall dispersion around the mean? If there was no dispersion, than all of the data points would equal the mean.

Two Points
 * Standard Deviation**
 * Summary description of how the data is dispersed around the mean.
 * Identifies the average distance from the mean, or the amount that the average score deviates from the mean.
 * If SD is greater than the mean of the sample, it is indicative of outliers in the data (may be significant - this is what the t-test is for?).
 * If the SD is large, great amount of variance in the data.


 * Assessment #2**: Grade Distribution Explained
 * Clarity
 * Identification of the different SD's indicated, and how that is interpreted / related back to the setting.


 * Describing Data** - When means are being compared, always include their standard deviations and the variable that is being studied. Probably a good idea to mention sample size as well (see Assignment #1).

For Second Weekend: t-Test and Review Notes