Please enable JavaScript.
Coggle requires JavaScript to display documents.
RESEARCH METHODOLOGY FOR MARKETING - Coggle Diagram
RESEARCH METHODOLOGY FOR MARKETING
DATA CLEANING
Consistency check
identify data that are oput of range because they are inadmissible and must be corrected
Missing responses
represents values of a variable that are unknown cause respondents did not answer correctly, provided ambiguos answers or the anwers were not properly recorded
BIVARIATE STATISTICS
this tipe of data involves two different variable. The analysis include causes and relationship. The main purpose is to find out the relationship among the two variables
Methods
to measure variability
STANDARD DEVIATION
it is the squared root of the variance
s = sq(s^2)
low: indicated that the data points tend to be the mean of the set
hig: mean that the data points are spread out over a wider range of values
RANGE
it is the fidderence of the maximum (highest) and minimum (lowest) scores in the distribution
QUARTILE
VARIANCE
It is the mean of the squared deviations
= ∑((x -Xm)^2)/n-1
to measure shape
SKEWNESS (ASYMMETRY)
It is the tendency of the deviation from the mean to be larger in one dirtection than the other
KURTOSIS
it is a measure of relatyive peakdness or flatness of the curve. normal distibution have kurtosis = 0. positive kurtosis = more peaked than normal distribution. negative = more flat than a normal distribution
to measure central location
MODE
(for nominal variables); it represents the highest peak of a distribution
MEAN
the mean is the sum of values of variables divided for the number of variables
MEDIAN
(for ordinal varibales)it is the central measure when the values are represented in ascending or descending order. When the number of variables is even, we make the mean of the two extremes
when we have a normal distribution, mode, mean and median have the same value
DESCRIPTIVE STATISTICS
It does not allow to draw any conclusion
INFERENTIAL STATISTCS
It allows to draw conclusion, to make inferences and to test hypothesis
UNIVARIATE STATISTICS
this type of data consists of only one variable. It does not deal with causes or relationship. the main purpose of the analysis is to describe data