Quantitative Research Methods for Social Scientists: Levels of Measurement, Central Tendency, Dispersion, Descriptive vs Inferential Statistics, SPSS Output Interpretation

This article discusses the four levels of measurement, measures of central tendency and dispersion, the difference between descriptive and inferential statistics, and the interpretation of SPSS output. It also defines key terms such as null hypothesis, independent and dependent variables, and extraneous variables. The subject is Quantitative Research Methods for Social Scientists, and the course code is LC572.

LC572 Quantitative Research Methods
for Social Scientist

TABLE OF CONTENTS
Presenting four level of measurement.........................................................................................1
Measure of Central tendency and Measure of Dispersion...........................................................1
Difference between Descriptive statistics and Inferential Statistics............................................1
Definition.....................................................................................................................................2
Interpretation of SPSS output......................................................................................................2
REFERENCES................................................................................................................................4

Presenting four level of measurement
The four level of measurement are as mentioned below:
 Nominal: In this, the data can be categorized and mainly used for qualitative labels. For
example, in a survey, it is mention about gender then it can be categorizing into two form
i.e. male and female.
 Ordinal: Under this, data is ranked from higher to lower like likert scale in which options
are provided in the form of strongly disagree, disagree, neither agree nor disagree, agree,
strongly agree (Marees and et.al., 2018).
 Interval: There is a space interval between each of the values in order to show the value.
For example, the temperature in Fahrenheit with a difference between 10 and 20 degrees.
 Ratio: The data used under this measurement ranked evenly spaced and also has a natural
zero (Salkind and Frey, 2021). For example, weight which can be in a zero however,
value of temperature cannot be measured under this, as it cannot be a zero.
Measure of Central tendency and Measure of Dispersion
Measure of central tendency is all about numbers that are tend to be cluster around the
middle of the values (Chakrabarty, 2021). In this, mean, mode and median are fall that describe
the different indication of a typical value distribution.
Measure of dispersion is estimated the normal value of a dataset which is also an
important part that helps to describe the spread of a data and variable towards a dataset. In this,
range, interquartile range, standard deviation and variance are fall that provide measure of
variability (Campbell, 2021).
Difference between Descriptive statistics and Inferential Statistics
Descriptive statistics is mainly concerned with describing a population under a study and
in this, the data is organised in such a well manner way that helps to determine the results.
However, under inferential statistics, observation is performed in order to draw conclusion in
better manner (Ali, Bhaskar and Sudheesh, 2019). Inferential test mainly compares, test and
predicts data and provide results in the form of probability. The main function of descriptive
statistics is to summarise the sample whereas inferential statistics try to reach in a conclusion by
examining about the population. That is why, it can be stated that descriptive statistics
1

summarise key features of a dataset whereas inferential allow to test hypothesis through a
broader population.
Definition
 Null Hypothesis and hypothesis: Alternative hypothesis is all about a difference between
two or more variables which is used by the researcher in order to observe the patter which
is not by chance (Guillén-Gámez and Ramos, 2021). On the other side, null hypothesis
reflects that there is no statistically different between the variable and that is why, the
hypothesis is not proved. So, it can be stated that the test always predicts no effect over
the dependent variable.
 Independent and dependent variable: Independent variable is considered as a cause
which is not affected from anything and it is not changed by any other variable which
researcher want to measure (Startup and Whittaker, 2021). However, dependent variable
is changes when there is a fluctuation identified from independent variable. For example,
the test score of a student is depend upon different factors which include how he/she
studied, sleep, revision etc. Here, dependent variable is test score and independent
variables are sleep, studied, revision (as it can be many)
 Extraneous Variable: In a quantitative data, it is a variable which a researcher is not
going to investigate but it has a potential affect upon the outcome of a research study
(Aityan, 2022). Therefore, if investigator left uncontrolled then it lead to inaccurate
conclusion between independent and dependent variables
Interpretation of SPSS output
Frequency table
Through the frequency table, it has been identified that out of 3146 member, 148 of them
stated that it is very common to have noisy neighbour with loud parties. Also, majority of them
(1635) of them stated that it is not at all common to have such noisy neighbour with loud parties.
Also, 1105 respondents stated that it is not very common to have such parties and 255 of them
stated that it is fairly common. Therefore, it can be stated that it is not actually common when
people have noisy neighbour with loud parties.
Crosstabulation
In accordance with the cross tab, it has been interpreted that out of 578 young adults, 38
of them stated that it is very common to have a noisy neighbour with loud parties and 67 of them
2

stated that it is fairly common and 235 of them stated that it is not very common whereas 238 of
them stated that it is not at all common to have noisy neighbour with loud parties. On the other
side, out of 1640 adults it has stated that 90 of them stated it very common whereas 134 of them
stated that it is fairly common and 608 of them reflected not very common and 808 of them
stated not at all common. However, out of 916 elderly people, 20 of them stated that it is very
common to have noisy neighbour and 134 of them stated that it is fairly common and 259 of
them stated not very common and 583 stated that not at all common. Therefore, it can be stated
that majority of them stated it is not at all common to have noisy neighbour with loud parties.
Chi-square results
In accordance with the asymptomatic significance P 0.00 it has been interpreted that there
is significant difference between the variable which means that views of selected respondents
varied from the noisy neighbour with loud parties. As a result, alterative hypothesis has been
accepted over null because the value of p is lower than standard criteria.
Phi/Cramer’s
The value of Cramer from the table entails that there is a lower association between the
variable and as per the phi value, it can be stated that 17% changes has been identified over the
dependent value. Therefore, it can be stated that there is minor changes identified over dependent
value though there is a significance difference between the variable.
3

REFERENCES
Books and Journals
Aityan, S. K., 2022. Introduction to Statistics. In Business Research Methodology (pp. 217-232).
Springer, Cham.
Ali, Z., Bhaskar, S. B. and Sudheesh, K., 2019. Descriptive statistics: Measures of central
tendency, dispersion, correlation and regression. Airway. 2(3). p.120.
Campbell, M. J., 2021. Statistics at square one. John Wiley & Sons.
Chakrabarty, D., 2021. Model Describing Central Tendency of Data. International Journal of
Advanced Research in Science, Engineering and Technology, pp.2350-0328.
Guillén-Gámez, F. D. and Ramos, M., 2021. Competency profile on the use of ICT resources by
Spanish music teachers: descriptive and inferential analyses with logistic regression to
detect significant predictors. Technology, Pedagogy and Education. 30(4). pp.511-523.
Marees, A.T. and et.al., 2018. A tutorial on conducting genome‐wide association studies: Quality
control and statistical analysis. International journal of methods in psychiatric
research. 27(2). p.e1608.
Salkind, N. J. and Frey, B. B., 2021. Statistics for people who (think they) hate statistics: Using
Microsoft Excel. Sage publications.
Startup, R. and Whittaker, E.T., 2021. Introducing social statistics. Routledge.
4

5