Research Methods- Chapter 13: Quantitative Data Analysis
The common error arises due to quantitative data analysis looks like a distinct phase that occurs after the
data have been collected.
U should be fully aware of what techniques u will apply at a fairly early stage for example when u r
designing ur questionnaire, observations schedule, coding frame, or whatever. The two main reasons for
this are as follows: 1) U can’t apply just any tech. to any variable. Tech. have to b appropriately match to
the types of variables that u have created thru ur research. This means that u must b fully conversant with
the ways in which diff types of variables are classified. 2) The size and nature of ur sample are likely to
impose the limitations on the kinds of tech u can use.
Simple Random sample-
Missing Data- An imp issue is how to deal with issue wen stuff are or info is missing. Missing Data arises
when respondents fail to reply to a question- either by accident or cuz they do not want to answer the
A type of Variables- one of the things that might strike u wen u look at the questions is that the kinds of
info that u receives varies by questions. The considerations lead to a classification of the different types of
variables that r generated in the courses of research. The four main types- 1) interval/ratio variables- These
r variables where the distances btw the categories r identical across the range of categories. The highest
level of measurements and a very wide range of tech of analysis can b applied to interval ratio variables.
There is in fact a distinction btw interval and ratio variables, in the the latter r interval variables with a fixed
zero point. However, most ratios variables exhibit this quality in social research they r not being
distinguished here. 2) Ordinal Variables- These r variables whose categories can b ranked ordered, but the
distances btw the categories r not equal across the range. 3) Nominal Variables- these variables also known
as categorical variables, comprise categories that cant b rank ordered. 4) Dichotomous Variables- These
variables contain data that have only two categories. Their position in relation to other types is slightly
ambiguous as they have only one interval. These therefore, can be considered as having attributes of the
other three types of variables. They look as though they r nominal variables, but cuz they have only one
interval they r sometimes treated as ordinal variables. However it is probably safest to treat them for most
purposes as if they were ordinary nominal variables.
Multiple – indicators (multiple items) measures of concepts like Likert scales strictly speaking produce
ordinal variables. Many writers argue that they can b treated as though they produce interval/ratio variables,
cuz of the relatively large number of categories they generate.
Univariate analysis- refers to the analysis of one variable at a time. 1) Frequency Tables- a frequency
table provides a number of ppl and the % belonging to each of the categories for the variables in question. It
can be used in relation to all for the different types of variables. If an interval/ratio variable is to b presented
in a frequency table format it is invariably the case that the categories will need to b grouped. 2) Diagrams-
r among the most frequently used methods of displaying quantitative data. Their chief adv. Is that they r
relatively easy to interpret and understand. If u r working with nominal or ordinal variables. The bar chart
or pie chart is two of the easiest methods to use. Each bar represents the number of ppl falling in each
category. Another way of displaying the same data is thru a pie chart. This also shows the relative size of
each slice relative size of the diffe