Website Opinions and Preferences Assignment

The assignment provides a set of questions related to website opinions and preferences, with students required to assume equal variances or not for statistical analysis. The questions cover various aspects such as the importance of websites, navigation, design, marketing strategies, user-friendliness, and costs associated with having a website. Students are expected to analyze data and draw conclusions based on the provided responses.

SPSS

ABSTRACT
Many survey studies uses questionnaire in which, respondents are asked individually to provide their responses about their
belief, attitude, demographic and intended future behaviours. The current research performed to gather the responses from 54
university students about the use of website access, its importance, innovation & creativity aspect and many others. To test the quality
of survey responses, reliability test through Cronbach alpha in SPSS has been performed. Moreover, Item analysis carried out to
examine the significance of difference in top and bottom 15% score for all the components. The findings of the study strongly evident
that the questionnaire had constructed so well and the results reliable also because Cronbach’s alpha statistics is 0.984 beyond 0.70.
Lastly, Item Statistics found significance level of mean difference in both the top and bottom 15% scoring from Q1 to Q10.

TABLE OF CONTENTS
ABSTRACT................................................................................................................................................................................................2
INTRODUCTION.......................................................................................................................................................................................1
METHODS..................................................................................................................................................................................................1
RESULTS....................................................................................................................................................................................................2
DISCUSSION..............................................................................................................................................................................................7
REFERENCES............................................................................................................................................................................................9
APPENDIX................................................................................................................................................................................................11
Cronbach’s alpha....................................................................................................................................................................................11
Independent sample T-test for Top and Bottom series..........................................................................................................................14

INTRODUCTION
The success of every research study depends on the sound knowledge of data collection techniques. As the current study
investigates website use therefore, a survey will be conduct that necessitates proficient knowledge of questionnaire construction. The
aim of the research is to investigate website users and assesses internal reliability of the responses provided by the respondents. In
order to check the reliability and validity test, statistical psychometric test of Cronbach alpha will be exercised in SPSS.
METHODS
Participants:
Investigator had designed questionnaire and used it as a research instrument to survey the selected people. Questionnaire is a
tool of primary data collection that contains a set of specifically designed questions to obtain statistically useful and meaningful
information by surveying people. The sample size is 54 and all were the fellow students who use website for different purpose. It
contains 10 questions to assess that how well students agree or disagree with different aspect such as use of website as a power tool of
communication, navigation, creativity and innovativeness, cost and other aspects.
Material and procedures
 Initially, questionnaire contains 10 questions that are 5-point Likert-scale questions comprising item from “strongly agree” to
“strongly disagree” had been designed and posted online on Survey Monkey. Questionnaire had been designed considering
readability of sentences so that respondent can easily understand the question in the same manner. In this, neither too short nor
too long sentences have been used and no technical wording had been used to avoid any confusion or lack of understanding
among the target people.
1

 Simplicity, ease to understand and use of clear language helps to ensure high response rate. At the time of drafting
questionnaire, negative sentences, technical terms and ambiguous statement has been avoided. Moreover, readability has been
checked using MS word. Moreover, cognitive interview has been conducted to compliment the readability check.
 Reliability simply means the quality or trustworthiness of the research outcome which is checked by developing standardized
questionnaire wherein students are asked with 10 different set of questions in same order. In this replication of the questions
creatively helps to check the consistency or reliability of the responses provided by people. In current research, internal
reliability of website user survey is checked through applying Cronbach’s alpha test in SPSS.
RESULTS
Psychometric analysis is a science that is used to produce the empirical evidence for the applied survey instrument. It includes
reliability aspect that is concerned with consistency, outcome precision and confidence of the research results or final outcome.
Cronbach’s alpha: In statistics; Cronbach’s alpha is the most useful way of measuring internal consistency. It is of huge significance
when there are multiple Likert scale questions (Bonett & Wright, 2015). As in the present study, researcher attempted to analyze the
use of website among students, therefore, in order to assess that questions comprised in the questionnaire reliably measure the latent
variable or not, this test can be used.
Output:
Demographic profile of participants
Appendix 1
Statistics
Age
N Valid 54
Missing 0
Mean 2.1667
2

Median 2.0000
Mode 2.00
Std. Deviation .69364
Variance .481
Age
Frequency Percent Valid Percent Cumulative Percent
Valid
Below 18 years 9 16.7 16.7 16.7
19 - 22 years 27 50.0 50.0 66.7
22 - 25 years 18 33.3 33.3 100.0
Total 54 100.0 100.0
3

According to the output, age statistics showed M = 2.16, SD = 0.69. It indicates that average age of the participants ranges
between 19-22 years. In the survey, 50% of the participants were under the age group of 19-22 years. However, 9 (16.67%) were
below 18 years and 18 (33.33%) of participants were in the age of 22-25 years.
Gender
Frequency Percent Valid Percent Cumulative Percent
Valid
Male 27 50.0 50.0 50.0
Female 27 50.0 50.0 100.0
Total 54 100.0 100.0
In the survey, male and female were in equal proportion because as per frequency statistics; 27(50%) were male and rest
27(50%) participants were female.
4

Cronbach’s alpha to test reliability aspect
Appendix 1
Case Processing Summary states that there are total 54 observation or responses and no dataset is excluded hence, the finding
can be easily generalized and does not seems problematic.
Cronbach’s alpha is a measurement of internal consistency among group of items from a single scale. With reference to the
designed questionnaire, “Reliability Statistics” presents the Cronbach’s alpha value is α=0.984 that evident relative greater internal
consistency because value above α =0.70 is always considered acceptable and justifiable (Trizano & Alvarado, 2016). It proves that
results are reliable and produce meaningful statistics about the website usage.
“Item Statistics” table presents mean and standard deviation scores for all the questions contained in the questionnaire. The
results highlighted the items with quite higher or lower value that needs to be removed to enhance reliability. As per the table, it can
be seen that Q1, Q7 and Q10 average score of (M =3.26 , SD=1.390),( M =3.06 , SD=1.036)∧( M=3.54 , SD =1.128) seems
comparatively higher than other items in the questionnaire. In order to decide whether these items should be removed or not, other
relevant statistical tables need to be checked (Tavakol & Dennick, 2011). On the other side, other questions mean prevails in similar
range from 2.04 to 2.70.
Table labeled as “Inter-Item Correlation Matrix”, it is found that all the identified components Q1, Q7 and Q10 with greater
mean score do not here found with weak correlation. The outcome clearly presents that these items are strongly relevant to other
questions with high value of correlation coefficient just like other items. Thus, on the basis of the results, it cannot be said that these
results are problematic (Gliem & Gliem, 2003).
From the table labeled as “Item Total Statistics”, it is clearly depicts that there are only question 1 and 6 which removal will
not bring any change in the value of Cronbach’s alpha whereas deletion of any other question lead to decline the reliability statistics
5

(Statistics, 2015). Q1 and Q6 alpha score is same to the Cronbach’s alpha and as per the Corrected Item-Total Correlation statistics,
both the questions had scored 0.910> 0.3 and 0.852>0.3 , therefore, there is no requirements of reversal scoring for these components.
There is no item exist which removal would tend to bring improvement in the Cronbach’s alpha (Goforth, 2015). Therefore, it is better
to not to remove any set of question in the questionnaire that can decrease the reliability statistics.
Compare means of two groups on each item from Q1 to Q10 through Independent Sample t-Test
Appendix 2.
Item discrimination is used to measure the effectiveness of an item. According to the “Group Statistics”, it can be seen that top
level under Q1 scored ¿) whereas the same for bottom series is determined with (M =0.88 , SD=0.354).Standard deviation that
measures the spreadness of a series value from its closest mean (Sedgwick, 2010). As per the results, it can be seen that Top series in
Q1 shows (M = 5.00, SD = 0.000) because all the values lies in top scoring is equal to 5 however, bottom score show ¿). On the
contrary, Q2 and Q8 determined (M =1 , SD =0.000)∧( M =1.00 , SD=0.000) for lowest 15% score and for upper 15% score, it is
(M =4.25 , SD=0.463)∧( M=1.00 , SD=0.00) respectively. Q10 states that all top 15% score are rated 5 at (M =5 , SD=0.000)
whereas bottom series evidenced ¿) respectively. On the other side, Q5, Q6 and Q7 found equal value of standard deviation for both
the top as well as bottom 15% score.
Looking to the table labelled as Independent Sample T-test for question 1, t (14)=33.000 , p=0.000<0.05 thus, as per the
results; it seems clear that there is significant level of mean difference exist in the Top & Bottom Score series ( Norušis, 2006).
Similarly, all the other questions ranging from Q2
found t (14)=19.858 , p=0.000<0.05 , for Q 3 t (14)=9.165 , p=0.000<0.05 , for Q 4 t(14 )=13.960 , p=0.000<0.05 , for Q5 , t (14)=18.385 , p=0.000<
.
In the entire situation, the value of significance is below the set alpha therefore, it becomes clear that alternative hypothesis
accepted which states significant statistical difference in the mean score of top and bottom series. Thus, the research results evident
significant level of statistical difference between the mean score of top and bottom series for every question from Q1 to Q10.
6

DISCUSSION
As per the analysis carried out, it becomes clear that the internal reliability measures found impressive result and assure that if
the same questionnaire is used again with a sample, then there is a strong possibility that surveyor will get the similar result. In other
words, researcher will obtain consistent results. As per the findings of Item total statistics, most of the questions designed in the
questionnaire are worth of retention because none of the item deletion is expected to enhance the reliability of the output. The findings
states that all the questions designed in the questionnaire are closely or strongly related to each other and none of the item deletion will
tend to increase the final outcome. Henceforth, it seems clear that constructed questionnaire is the designed so well to survey all the
website users.
In the outcome, none of the Cronbach’s alpha score is found negative. Moreover, although two components Q1 and Q6 had
scored the same Cronbach’s alpha and Corrected Item Total Correlation is beyond the threshold point of 0.3. Therefore, the
components were not recognised to be requiring for reverse scoring. According to the Item analysis, significant level of statistical
difference discovered among all the components for Top and Bottom scoring because of zero sig level which is below the alpha of
0.05.
Questionnaire that is used in the research as a research instrument has many strengths as well as drawbacks. One of the most
effective strength of the method is it is the most affordable way. Referring from the current study, as online questionnaire has been
used and responses were gathered through survey monkey, therefore, it does not incur too much expense hence considered as a cost-
efficient way of survey. Apart from this, as responses are gathered from a 54 university students who often access different websites
for varied purpose, therefore, it is found as a practical way (Debois, 2016). Despite this, online survey did not require excessive time
in data gathering henceforth, speedy responses were collected. Apart from this, 100% response rate was the key strength of the
questionnaire survey.
However, on the other side, there was an exception because for 4 questions, 1 candidate did not provide their responses.
Moreover, many-times, it may be a possibility that respondents do not truthfully answer the questions, hence, dishonesty and lack of
7

conscientious response are the major drawback (De Franzo, 2012). With reference to the current study, as Cronbach alpha measure
was used to test the reliability aspect and the findings evidence that there the responses are reliable and trustworthy because the alpha
statistics is measured beyond 0.70.
8

REFERENCES
Books and Journals
Bonett, D. G., & Wright, T. A. (2015). Cronbach's alpha reliability: Interval estimation, hypothesis testing, and sample size
planning. Journal of Organizational Behavior. 36(1). 3-15.
Gliem, J. A., & Gliem, R. R. (2003). Calculating, interpreting, and reporting Cronbach’s alpha reliability coefficient for Likert-type
scales. Midwest Research-to-Practice Conference in Adult, Continuing, and Community Education.
Goforth, C. (2015). Using and interpreting Cronbach’s Alpha. Statistical Consulting Associate, University of Virginia Library.
Norušis, M. J. (2006). SPSS 14.0 guide to data analysis. Upper Saddle River, NJ: Prentice Hall.
Sedgwick, P., 2010. Independent samples t test. BMJ: British Medical Journal (Online). 340.
Statistics, L., 2015. Cronbach's alpha using SPSS Statistics. Statistical tutorials and software guides.
Tavakol, M., & Dennick, R. (2011). Making sense of Cronbach's alpha. International journal of medical education, 2, 53.
Trizano-Hermosilla, I., & Alvarado, J. M. (2016). Best alternatives to Cronbach's alpha reliability in realistic conditions: congeneric
and asymmetrical measurements. Frontiers in psychology. 7. 769.
Online
De Franzo, S. (2012). Advantage and disadvantages of survey. [Online]. Available through: <
https://www.snapsurveys.com/blog/advantages-disadvantages-surveys/>.
Debois, S. (2016). 9 Advantage and disadvantages of Questionnaires. [Online]. Available through: <
https://surveyanyplace.com/questionnaire-pros-and-cons/>.
9

10

APPENDIX
Cronbach’s alpha
Case Processing Summary
N %
Cases
Valid 54 100.0
Excludeda 0 .0
Total 54 100.0
a. Listwise deletion based on all variables in the procedure.
Reliability Statistics
Cronbach's Alpha Cronbach's Alpha Based on Standardized
Items
N of Items
.984 .985 10
Item Statistics
Mean Std. Deviation N
Q1 Everybody needs a website 3.26 1.390 54
Q2 Websites are one of the most powerful tools used
for communication. 2.54 1.145 54
Q3 Most websites are easy to navigate. 2.70 1.143 54
Q4 People prefer more colorful websites. 2.54 .985 54
Q5 Websites are the best way to generate and promote
your business. 2.50 1.060 54
Q6 Websites are generally user friendly. 2.46 .926 54
Q7 Websites are the best form of marketing. 3.06 1.036 54
11

Q8 Websites should be creative and innovative. 2.04 .951 54
Q9 All successful businesses have a website. 2.35 1.231 54
Q10 It costs a lot of money to have a website. 3.54 1.128 54
Inter-Item Correlation Matrix
Q1
Everybody
needs a
website
Q2
Websites
are one of
the most
powerful
tools used
for
communica
tion.
Q3 Most
websites
are easy
to
navigate
.
Q4
People
prefer
more
colorful
websites
.
Q5
Websites
are the
best way
to
generate
and
promote
your
business.
Q6
Websites
are
generally
user
friendly.
Q7
Websites
are the
best form
of
marketing
.
Q8
Website
s should
be
creative
and
innovati
ve.
Q9 All
success
ful
busines
ses
have a
websit
e.
Q10 It
costs a
lot of
money
to have
a
website
.
Q1 Everybody needs a
website 1.000 .883 .786 .847 .897 .697 .908 .863 .872 .957
Q2 Websites are one
of the most powerful
tools used for
communication.
.883 1.000 .888 .928 .941 .829 .898 .900 .921 .883
Q3 Most websites are
easy to navigate. .786 .888 1.000 .915 .904 .881 .891 .843 .920 .784
Q4 People prefer more
colorful websites. .847 .928 .915 1.000 .967 .860 .877 .865 .915 .839
12

Q5 Websites are the
best way to generate
and promote your
business.
.897 .941 .904 .967 1.000 .837 .886 .861 .919 .876
Q6 Websites are
generally user
friendly.
.697 .829 .881 .860 .837 1.000 .740 .859 .881 .715
Q7 Websites are the
best form of
marketing.
.908 .898 .891 .877 .886 .740 1.000 .860 .902 .895
Q8 Websites should be
creative and
innovative.
.863 .900 .843 .865 .861 .859 .860 1.000 .924 .843
Q9 All successful
businesses have a
website.
.872 .921 .920 .915 .919 .881 .902 .924 1.000 .867
Q10 It costs a lot of
money to have a
website.
.957 .883 .784 .839 .876 .715 .895 .843 .867 1.000
Item-Total Statistics
Scale Mean if
Item Deleted
Scale Variance if
Item Deleted
Corrected Item-
Total Correlation
Squared Multiple
Correlation
Cronbach's Alpha
if Item Deleted
Q1 Everybody needs a website 23.72 82.204 .910 .956 .984
Q2 Websites are one of the
most powerful tools used for
communication.
24.44 85.535 .956 .927 .981
13

Q3 Most websites are easy to
navigate. 24.28 86.280 .918 .933 .982
Q4 People prefer more colorful
websites. 24.44 88.553 .947 .951 .982
Q5 Websites are the best way to
generate and promote your
business.
24.48 87.009 .958 .967 .981
Q6 Websites are generally user
friendly. 24.52 91.160 .852 .892 .984
Q7 Websites are the best form
of marketing. 23.93 87.919 .931 .929 .982
Q8 Websites should be creative
and innovative. 24.94 89.563 .923 .912 .982
Q9 All successful businesses
have a website. 24.63 83.898 .961 .938 .981
Q10 It costs a lot of money to
have a website. 23.44 86.742 .907 .929 .982
Scale Statistics
Mean Variance Std. Deviation N of Items
26.98 107.075 10.348 10
Independent sample T-test for Top and Bottom series
Group Statistics
Group N Mean Std. Deviation Std. Error Mean
Q1 Everybody needs a website Top 8 5.00 .000 .000
Bottom 8 .88 .354 .125
Q2 Websites are one of the most Top 8 4.25 .463 .164
14

powerful tools used for
communication. Bottom 8 1.00 .000 .000
Q3 Most websites are easy to
navigate.
Top 8 4.50 .535 .189
Bottom 8 1.50 .756 .267
Q4 People prefer more colorful
websites.
Top 8 4.13 .354 .125
Bottom 8 1.25 .463 .164
Q5 Websites are the best way to
generate and promote your business.
Top 8 4.13 .354 .125
Bottom 8 .88 .354 .125
Q6 Websites are generally user
friendly.
Top 8 4.25 .463 .164
Bottom 8 1.75 .463 .164
Q7 Websites are the best form of
marketing.
Top 8 4.38 .518 .183
Bottom 8 1.63 .518 .183
Q8 Websites should be creative and
innovative.
Top 8 3.75 .463 .164
Bottom 8 1.00 .000 .000
Q9 All successful businesses have a
website.
Top 8 4.50 .535 .189
Bottom 8 .88 .354 .125
Q10 It costs a lot of money to have a
website.
Top 8 5.00 .000 .000
Bottom 8 1.63 .518 .183
Independent Samples Test
Levene's Test for
Equality of
Variances
t-test for Equality of Means
F Sig. t df Sig. (2-
tailed)
Mean
Differenc
e
Std. Error
Differenc
e
95% Confidence
Interval of the
Difference
Lower Upper
15

Q1 Everybody
needs a website
Equal variances
assumed 5.444 .035 33.00
0 14 .000 4.125 .125 3.857 4.393
Equal variances
not assumed
33.00
0 7.000 .000 4.125 .125 3.829 4.421
Q2 Websites are
one of the most
powerful tools
used for
communication.
Equal variances
assumed 21.000 .000 19.85
8 14 .000 3.250 .164 2.899 3.601
Equal variances
not assumed
19.85
8 7.000 .000 3.250 .164 2.863 3.637
Q3 Most
websites are easy
to navigate.
Equal variances
assumed 1.000 .334 9.165 14 .000 3.000 .327 2.298 3.702
Equal variances
not assumed 9.165 12.60
0 .000 3.000 .327 2.291 3.709
Q4 People prefer
more colorful
websites.
Equal variances
assumed 1.577 .230 13.96
0 14 .000 2.875 .206 2.433 3.317
Equal variances
not assumed
13.96
0
13.09
3 .000 2.875 .206 2.430 3.320
Q5 Websites are
the best way to
generate and
promote your
business.
Equal variances
assumed .000 1.000 18.38
5 14 .000 3.250 .177 2.871 3.629
Equal variances
not assumed
18.38
5
14.00
0 .000 3.250 .177 2.871 3.629
Q6 Websites are
generally user
friendly.
Equal variances
assumed .000 1.000 10.80
1 14 .000 2.500 .231 2.004 2.996
Equal variances
not assumed
10.80
1
14.00
0 .000 2.500 .231 2.004 2.996
Q7 Websites are
the best form of
Equal variances
assumed .000 1.000 10.62
7 14 .000 2.750 .259 2.195 3.305
16

marketing. Equal variances
not assumed
10.62
7
14.00
0 .000 2.750 .259 2.195 3.305
Q8 Websites
should be
creative and
innovative.
Equal variances
assumed 21.000 .000 16.80
3 14 .000 2.750 .164 2.399 3.101
Equal variances
not assumed
16.80
3 7.000 .000 2.750 .164 2.363 3.137
Q9 All
successful
businesses have a
website.
Equal variances
assumed 9.000 .010 15.99
9 14 .000 3.625 .227 3.139 4.111
Equal variances
not assumed
15.99
9
12.14
1 .000 3.625 .227 3.132 4.118
Q10 It costs a lot
of money to have
a website.
Equal variances
assumed 105.000 .000 18.44
5 14 .000 3.375 .183 2.983 3.767
Equal variances
not assumed
18.44
5 7.000 .000 3.375 .183 2.942 3.808
17