-
Descriptive statistics
methods of organizing, summarizing, and presenting data in informative way. Not estimating future populatio etc but only to summarize pst (i.e. canadian population over years)
-
Inferential statistics
methods used to estimate a property of a population using asample
Best guess of a population value based on a sample informationimpossible to test all of the population, so representative sample chosen
-
Qualitative data
identifies categories or attributes. Eg. colour, gender etc.
-
Quantitative data
has numerical attributes. It indicates how much or how many.
-
Discrete variables
can only assume certain values, and there are gaps between variables.(number of bedrooms)
-
Continuous variables
assume any variable within a specific range. (tire pressure)
-
At nominal level of measurement
observations of a qualitative variable can only beclassified and counted. There is no particular order to the labels. (colour of M&M’s, gender)
-
The properties of ordinal level
data are that data classifications are represented by sets oflabels that have relative values and can be ranked or ordered. Distance between valuesnon necessarily equal.
-
In Interval level data,
responses classify, indicate order, and the distances betweenconsecutive numbers are meaningful.Data are always numericalZero is a point on a scale – not necessarily the absence of a phenomenon or bottomof the scale
-
Ratio data
Interval data + absolute zero and the ratio of two number is meaningfulIn this case an absolute zero means the zero value represents the absence of thecharacteristic being studied.
-
Inferential statistics
methods used to estimate a property of a population using a sample.
-
Frequency table
a grouping of qualitative data into mutually exclusive classes showing thenumber of observations in each class.
-
Relative (class) frequencies
indicate the fraction of observations in each category (class)
-
A relative frequency distribution
is the decimal expression of the relative frequency, forall observations
-
Frequency distribution
a grouping of data into mutually exclusive classes showing the numberof observations in each class.1. Decide on the number of classes. 2 to the k rule. 2^k>n (k classes, n observations), e.g.2^k>40, k>ln(40)/ln(2), k>5.322, set k=62. Determine the class interval or width. �≥�−�/� where i=interval, H=max, L=min3. Set the individual class limits. Avoid overlapping or unclear class limits 1<=Rate<34. Tally the list prices into the classes.5. Count the number of items in each class. The number of observations in each class iscalled the class frequency.
-
Histogram
Like a bar chart, but used for quantitative dataHorizontal axis is continuous if data continuousBars are adjacentClass frequencies on vertical axis, class limits OR class midpoints on horizontal axis.
-
A frequency polygon
also shows the shape of a distribution and is similar to a histogram. Itconsists of line segments connecting the points formed by the intersections of the classmidpoints and the class frequencies.
|
|