Quantitative

Research
Methods
Business and Management
Research

By: Janus Aries Q. Simbillo

What is Quantitative
Research?
 Quantitative methods emphasize on objective
measurements and numerical analysis of data collected
through polls, questionnaires or surveys. Quantitative
research focuses on gathering numerical data and
generalizing it across groups of people.
 According from Aliaga and Gunderson (2000),
Quantitative Research is ‘Explaining phenomena by
collecting numerical data that are analyzed using
mathematically based methods (in particular statistics)’.
 The goal is to determine the relationship between one
thing (an independent variable) and another (a
dependent or outcome variable) in a population.
Main Characteristics of
Quantitative Research
 The data is usually gathered using more structured research
instruments.
 The results are based on larger sample sizes that are representative of
the population.
 The research study can usually be replicated or repeated, given its high
reliability.
 Researcher has a clearly defined research question to which objective
answers are sought.
 All aspects of the study are carefully designed before data is collected.
 Data are in the form of numbers and statistics.
 Project can be used to generalize concepts more widely, predict future
results, or investigate causal relationships.
 Researcher uses tools, such as questionnaires or equipment to collect
numerical data.
Different types of data for
analysis
 Internal Variables
 The highest form of measurement and the
easiest to manipulate and analyze.
 There is a fixed space (interval) between
each variable and this is a consistent
space.
 Answers involving age, income, number of
staff, revenue, etc. (items that are highly
measurable)
Different types of data for
analysis
 Ordinal Variables
 These can be rank ordered but the space
between the variables is not equal across the
range.
 These categorize the interval variables, thus
making the data more useful for further analysis.
 Nominal Variables
 These variables cannot be rank ordered at all.
 These are more descriptive in nature rather than
numerical
Different types of data for
analysis
 Dichotomous Variables
 These are answers which can only fall into
one or two categories.
 Usually these are a yes/no answer or a
male/female gender
 Treated as a special kind of nominal
variable
Coding and Entering Data for
Computer Analysis
 Data Matrix
 Sets out the data in a matrix.
 Can be done with Excel or any other Spreadsheet based
program; or can be also done through the program,
SPSS for Windows.
 Answers will need coding
 Coding
 Converting different responses into codes
 2 reasons why this is vital
 Codes are often worked out on scraps of paper quite quickly
 Will be having a hard time to remember exactly what the
results mean
Coding and Entering Data for
Computer Analysis
 Using SPSS for Windows
 SPSS stands for “Statistical Package for
Social Sciences”
 It is the most commonly used tool to produce
all statistical tests and analysis outlined.
 Weighting cases
 This is possible when using stratified random
sampling and when there is an unequal
response rate for different strata.
Presentation of Findings
 Frequency tables – univariate
 Shows a list of categories (types of response)
and the numbers of people responding to each.
 Categorize the responses to different groups for
more summarized information to be presented
 Bar Charts, Histograms and Pie Charts –
univariate
 Generally used for nominal or ordinal variables.
 For Interval Variables, a histogram would be
used rather than a bar chart.
Presentation of Findings
 Measures of Central Tendency –
univariate
 Mean, Median or Mode in a range of values
 Not representable in a chart but in a series
of means, medians, etc., this can be
charted or shown in a table.
 Mean is for Interval Variables only, Median
is for both Interval and Ordinal Variables,
Mode can be calculated for any variable.
Presentation of Findings
 Measures of dispersion – univariate
 Range, Inter-quartile Range or Standard
Deviation
 Box Plot can be used for both Central
Tendency and Dispersion in a chart format.
 Charts, diagrams and tables – the detail
 Trends over time
 Usually shown in a line graph where time
is on the horizontal axis.
Selecting Appropriate
Statistical Tools
 Relationships between variables – bivariate
analysis
 Means the variation in one variable coincides with
variation in another variable.
 It does not necessarily follow that one will be an
independent and one a dependent variable.
 Contingency tables or cross-tabulations –
bivariate
 Set up as a frequency table including column
percentages but showing both variables against
the chosen categories
Selecting Appropriate
Statistical Tools
 Pearson’s r (Pearson’s Product Moment Correlation
Coefficient – PMCC) - bivariate
 Looks for relationships between interval variables.
 Between 0 to 1, so the closer the Pearson’s r is to 1,
the stronger the relationship between 2 variables.
 Either positive or negative, which indicates the
direction of the relationship.
 Before calculating for Pearson’s r, consider to
construct a scatter diagram for the two variables, as
it should only be used when there is a broader linear
relationship, it will not hold for a curve relationship.
Selecting Appropriate
Statistical Tools
 Regression Analysis
 A coefficient of determination/regressions
coefficient can be calculated by squaring the
value of Pearson’s r and multiplying it to 100.
 It describes the proportion of variation in one
dependent variable accounted for by other
independent variable.
 When a analysis involves multiple
independent variables, it is called multiple
regression analysis
Selecting Appropriate
Statistical Tools
 Spearman’s rho (p) – bivariate
 Used when at least one of the 2 variables is ordinal and the other
is either ordinal or interval.
 Same outcome as Pearson’s r
 Phi and Cramer’s V – bivariate
 Phi is used to explore the relationship between 2 dichotomous
variables while Cramer’s V is used for the relationship between 2
nominal variables.
 Phi outcomes are like Pearson’s r or Spearman’s rho but for
Cramer’s V only the strength of the relationship can be shown.
 Eta – bivariate
 Used to explore relationships between an interval variable and a
nominal variable but only shows the strength of the relationship
and not the direction.
Selecting Appropriate
Statistical Tools
 Statistical significance
 A way of testing the level of confidence using a
probability sampling technique that has generated
results that can be applied to the full population
 Can also estimate the chances of no relationship (null
hypothesis) in fact existing between two variables,
when bivariate analysis suggests that there is.
 Setup first the null hypothesis, then decide the level of
statistical significance that can be accepted (i.e. the
level of risk that the null hypothesis will be rejected)
 Use Chi-square test (x2) to produce the level of
significance (p or probability level)
Selecting Appropriate
Statistical Tools
 Testing whether groups are different – multivariate
analysis
 Independent groups t-test can be used to test the
probability of the groups being different, when a
quantifiable variable can be split into two groups using
a descriptive variable. The lower the t statistic, the
more likelihood of any difference in the groups being
caused by chance.
 Paired t-test can be used to measure pairs of variables.
 ANOVA or f-test, when testing the differences between
three or more groups to see if there is really a
“statistically significant difference”.
THANK YOU!
Have a nice day ahead!