Rapple+Journal+4

Back to Lisa's Journal or Go to Journal 5

3 DVB Frequency tables 4 DVB IQR/Median 6 DVB Standard deviation/Mean


 * Frequency distribution:** is the frequency that a certain variable occurs in the sample group. If you take a histogram and draw a line over the upper limits it represents frequency distribution.

The shape of the data will determine if you use parametrics (normal distribution only) or non-parametrics. The non-parametrics will recalculate to approximate.
 * CENTRAL TENDENCY: You can evaluate or represent the central tendency of the data in three ways:**
 * 1) mode (mo) - can be calculated for nominal, ordinal, or continuous data. The mode represents the most frequently occurring score in the data. It can be bimodal, whereas the distribution has two peaks that are either equal in height or //approximately// the same height as one another. Multimodal means there are three or more peaks on the frequency distribution histogram.
 * 2) median (mdn) - can be calculated for ordinal or continuous data only. The median score represents the middle point in a set of ranked scores or measures. The value of each score is not considered except to place it on the frequency distribution. Then the scores are divided in half so that half fall on one side and half fall on the other side of the MEDIAN score.
 * 3) mean ( //x// ) - can be calculated for continuous data only. Because mean takes into account the arithmetic value of the data, the data can only be continuous (or scale) where it has a mathematical value. All the data values are added together and then divided by the total number of datum to get an average value.
 * Shape of Data**: the dispersion, or range, of the data values will be reflected in the shape of the data. A normal distribution, a bell curve, means that the mode, median, and mean are all equal. If there are more outliers to the right of the distribution curve than the bell curve is said to be "skewed" to the right, and the same can be be said for outliers to the left. This may be expressed as "skewness".

Tables:
 * Frequency table (becomes a "relative" table if use %/proportions)
 * Contingency table (have two categorical variables)

Distributions: Independent variables - if there is no association between the variables, they are independent of one another. The contingency table/bar chart/etc. will show the same distribution for all categories.
 * Marginal distribution: the frequency distribution (%) of one variable.
 * Conditional distribution: considers a smaller group within the study. answers a "contingent" question. "Was X contingent upon Y?"

Dependent variables - Find out by asking if the variables are NOT associated. 3 DVB bar charts, pie charts (qual) 4 DVB Stem & Leaf, Histograms 5 DVB Boxplots, Timeplots