Professional Research and Communication

This article discusses the use of Likert-type scale, appropriateness of collected data, classification of data types and test of hypothesis for online and classroom education.

Running head: PROFESSIONAL RESEARCH AND COMMUNICATION
Professional Research and Communication
Name of the Student:
Name of the University:
Author Note:

1PROFESSIONAL RESEARCH AND COMMUNICATION
Table of Contents
Response to Question 1...................................................................................................................2
Response to Question 2...................................................................................................................2
Response to Question 3...................................................................................................................3
Response to Question 4...................................................................................................................4
References........................................................................................................................................7

2PROFESSIONAL RESEARCH AND COMMUNICATION
Response to Question 1
The given survey uses responses of the respondents based on the Likert-type scale which
is a measurement device to measure psychological opinions. The responses are distributed in 5
categories which are –
1 for Strongly Disagree, 2 for Disagree, 3 for Unsure, 4 for Agree, and 5 for Strongly
Agree. Now, the number of responses of each category is multiplied with the corresponding
numbers assigned to it and the single score is evaluated as 3.19. The procedure for evaluating
this single score cannot considered as an optimal solution. This output value 3.19 belongs
between two different categories “Unsure” and “Agree”. Moreover, this average score value
leads to assume that the survey reveals a mixture of neutral and agreement opinions. The dataset
is an ordinal dataset and the given problem tries to calculate the mean value of an ordinal dataset.
However, the mean, even the weighted mean cannot be considered as a measure of central
tendency for any ordinal data since on cannot take average of “Neutral” response and “Agree”
response. Moreover, it is completely meaningless to try to find out the average of “Strongly
Agree” and “Disagree” responses. The above result 3.19 does not interpret anything and one
cannot conclude any statistical interpretation form this result. In addition to this, the mean value
will represent nothing but a distortion from the collected psychological responses
(Measuringu.com, 2018).
The best statistical measure in this context would be a graphical representation of the
dataset using a bat chart.

3PROFESSIONAL RESEARCH AND COMMUNICATION
Response to Question 2
No, the collected data does not provide an appropriate reflection of the desired outcome
of the election. The procedure of collecting data by the polling company is to ask every passers-
by of the street corners in the capital city on week-days. These passers-by are not the
representatives of the entire population of voters in the capital city as theses passers-by may not
be residents of the capital city. They may be residents of other places. Moreover, these passers-
by are not chosen randomly and as a result, the statistical measure will not be significant. Apart
from that, the responses coming from the passers-by of the street corners do not constitute cross-
sectional data. Thus, the dataset does not reflect the true picture of the targeted population. In
addition to this, the collected data using the above mentioned method will not constitute a
statistical random sample and the statistical measures will contain bias. No probability will be
associated with the units of the population to be selected into the sample (Acharya et al., 2013).
These points justify the answer.
Response to Question 3
There are five types of data and they are to be classified under Nominal, Interval, Ordinal,
and Ratio types of data. Firstly, these four quantitative data types are defined here and then the
five examples are classified under the four data types.
Nominal data are those data which cannot be arranged into any order and can only be
allocated to different categories. Examples of this named data are gender, hair color and many
more (Allison, 2014).

4PROFESSIONAL RESEARCH AND COMMUNICATION
Ordinal data, as the name suggests, can be arranged into certain order. For examples, rank of
students in any competition and many more. These type of data are described on a rating scale of
1 to 10.
Interval data scales provide measurements on a scale where the difference between two
points on the scale are exactly equal and there is no meaningful absolute zero point. Temperature
is good example of interval data (Freelon, 2013).
Lastly, ratio data scale is almost similar to the interval scale but it has a meaningful zero
point along with other numeric values on the scale (LoBiondo-Wood et al., 2013). This true zero
point ensures no possibility of having negative values on the scale. Distance moved by a
projectile (Freelon, 2013).
a. The number of cars passing through an intersection in an hour, in whole numbers – this is
the example of interval scale as it is calculating frequency (Murray, 2013).
b. The temperatures measured on Kelvin scale can be classified under the class of ratio scale
as Kelvin thermometer has an absolute zero point. This is why a temperature of 200
Kelvin is twice high than the temperature of 100 Kelvin.
c. Fahrenheit thermometers measure the temperature on interval scale as the difference
between two temperatures is measurable and the zero point is random.
d. The type of mobile phones possessed by anyone is considered as Nominal data as the
brand of the mobile phone (Nokia, Samsung, Apple or others) cannot be arranged in any
order, they are named data (Li, 2013).
e. A person’s height is measured on a ratio scale as it measures exact difference value
between two quantities and there is an absolute zero point in the scale. Moreover, the
height cannot be negative. Thus, it is an example of ratio data.

5PROFESSIONAL RESEARCH AND COMMUNICATION
Response to Question 4
In this question, a test of hypothesis needs to be performed based on three different
studies. The problem requires to assess the student’s performance for both online education and
classroom education to conclude which mode education is more likely to help improving the
student’s performance. The test under the three different studies are describes below-
a. Descriptive non-experimental study
The non-experimental study does not require manipulation of the independent
variables of the study. Moreover, predictor variables are not controlled by the researcher
and they are concluded on the basis of interpretation, interaction and observation.
Therefore, to check the students’ performances, the researcher cannot construct a cause-
and-effect relationship and has to proceed with the case studies and correlation to
measure the strength of association. The Non-experimental research has chances for high
level of variability. The research hypothesis for Non-experimental research can be
defined on a single variable that is the test score of students in this case for the two types
of education procedures. The participants cannot be assigned randomly in this type of
experiments (Bleske-Rechek, Morrison & Heidtke, 2015).
b. Quasi experimental study
The researcher can control the independent variables but the observations of the study
cannot be randomly assigned in the Quasi-experimental research method. It is called
Quasi-experimental because the assignment of the variables into the experimental groups
is intentional and non-random. To evaluate the measure of improvement of the students’
performances, the researcher may like to perform the test by comparing the test scores for

6PROFESSIONAL RESEARCH AND COMMUNICATION
the two groups – group of the students taking online classes and group of students
attending the classroom program. The outcome of this comparison will conclude which
educational program will help to improve the students’ performances (Campbell &
Stanley, 2015).
c. Experimental study
The experimental design gives the scope to the researcher to control the subjects and
prepare a cause-and-effect relationship. The manipulation of the experiment will yield the
outcome the test. This type of study is more of a lab-based experiment and the test is
performed in any lab (Harpe, 2015). It is the best method among the three to assess the
causal hypothesis. The given problem can be tested in the true experimental study by
performing a statistical t-test. The variable will be the test scores of the two groups of the
students taking classes online and in classroom.

7PROFESSIONAL RESEARCH AND COMMUNICATION
References
Acharya, A. S., Prakash, A., Saxena, P., & Nigam, A. (2013). Sampling: Why and how of it.
Indian Journal of Medical Specialities, 4(2), 330-333.
Allison, P. D. (2014). Event history and survival analysis: Regression for longitudinal event
data (Vol. 46). SAGE publications.
Bleske-Rechek, A., Morrison, K. M., & Heidtke, L. D. (2015). Causal inference from
descriptions of experimental and non-experimental research: Public understanding of
correlation-versus-causation. The Journal of general psychology, 142(1), 48-70.
Campbell, D. T., & Stanley, J. C. (2015). Experimental and quasi-experimental designs for
research. Ravenio Books.
Freelon, D. (2013). ReCal OIR: Ordinal, Interval, and Ratio Intercoder Reliability as a Web
Service. International Journal of Internet Science, 8(1).
Harpe, S. E. (2015). How to analyze Likert and other rating scale data. Currents in Pharmacy
Teaching and Learning, 7(6), 836-850.
Li, Q. (2013). A novel Likert scale based on fuzzy sets theory. Expert Systems with Applications,
40(5), 1609-1618.
LoBiondo-Wood, G., Haber, J., Berry, C., & Yost, J. (2013). Study Guide for Nursing Research-
E-Book: Methods and Critical Appraisal for Evidence-Based Practice. Elsevier Health
Sciences.
Measuringu.com, (2018) MeasuringU: Can You Take the Mean of Ordinal Data? . Retrieved 12
August 2018, from https://measuringu.com/mean-ordinal/

8PROFESSIONAL RESEARCH AND COMMUNICATION
Murray, J. (2013). Likert data: what to use, parametric or non-parametric?. International Journal
of Business and Social Science, 4(11).