You are on page 1of 6

RQ1: Demographic profile of business trading with the fish farm

1b) A test that the proportion of all customers with a business life of less than a year is equal to 20%

H0: The %of business life of less than a year =20% H1: The % of business life of less than a year <> 20%

Two tail test Pvalue = 0.475* 2= 0.95 Level of sig= 0.05 Conclusion Pvalue > Level of significance; We
do not reject the Null; there is a strong evidence for the null at 5% level of significance; So, it implies the
% of trading business having age less than 1 year is not significantly different than 20%; the claim is valid

1c) A test that the proportion of all customers that operate as “Owner only” is equal to 10%

H0: The % of Business that operates as “Owner only” = 10% H1: The % of Business that operates as
“Owner only” <> 10%

Two tail test Pvalue = 0.248* 2=0.496 Level of significance= 0.05 Conclusion Pvalue > Level of
significance; We do not reject the Null; There is strong evidence for the null at 5% level of significance;
So, it implies the % of business that operates as owner only is not significantly different than 10%; the
claim is valid

1d) A test that the proportion of customers with an income of €150k or more is equal to 25%

H0: The % of Business having annual income of £150k or more = 25% H1: The % of Business having
annual income of £150k or more <> 25% Two tail test Pvalue = 0.098* 2= 0.196 Level of significance=
0.05 Conclusion Pvalue > Level of significance; We do not reject the Null; There is a strong evidence for
the null at 5% level of significance; So, it implies that % of the customers with an income of £150k or
more is not significantly different than 25%; the claim is valid

RQ2:

2c) A test that the average score for each service quality factor is different to 12

For Responsiveness

H0: The average score for responsiveness service quality factor = 12 H1: The average score for
responsiveness service quality factor <> 12 Two Tail test Pvalue= 0.077 Level of significance = 0.05

Conclusion Pvalue > Level of significance; We do not reject the null hypothesis; There is strong evidence
for the null hypothesis at 5% level of significance; It implies that the average score for responsiveness
service quality factor is not significantly different to 12; The claim is invalid

For Delivery

H0: The average score for delivery service quality factor = 12 H1: The average score for delivery service
quality factor <> 12 Two Tail test Pvalue= 0.479 Level of significance = 0.05 Conclusion Pvalue > Level of
significance;We do not reject the null hypothesis; There is strong evidence for the null hypothesis at 5%
level of significance; It implies that the average score for delivery service quality factor is not
significantly different to 12; The claim is invalid

For Tangibles

H0: The average score for tangibles service quality factor = 12; H1: The average score for tangibles
service quality factor <> 12 Two Tail test Pvalue= 0.003 Level of significance = 0.05 Conclusion
Pvalue <Level of significance; We reject the null hypothesis; There is strong evidence against the null
hypothesis at 5% level of significance; It implies that the average score for responsiveness service quality
factor is significantly different to 12 .The claim is valid

2d) A test of significant difference in the average of each service quality factor and their base.

For Responsiveness

H0: Average of responsiveness = Average base of business H1: Average of responsiveness ≠ Average
base of business

Two Tail test Levene’s Test Pvalue=0.001 Level of significance= 5% (0.05) Pvalue < Level of significance
Equal variance not assumed

Equality of Mean Test

Two Tail test Pvalue= 0.00 Level of Significance= 5% (0.05)

Conclusion Pvalue < Level of significance; We reject the null; There is strong evidence against the null at
5% level of significance with equal variance not assumed

It implies that there is significant difference between average service quality for responsiveness and
average base of the business( OR; it implies that the avg score for responsiveness & base are
significantly different). The claim is valid

For Delivery

H0: Average service quality for delivery = Average base of business H1: Average service quality for
delivery≠ Average base of business

Two tail test Levene’s Test Pvalue=0.635 Level of significance= 5% (0.05)

Pvalue > Level of significance Equal variance assumed

Equality of Mean Test Two tail test Pvalue= 0.00 Level of Significance= 5% (0.05)

Conclusion Pvalue< Level of significance; We reject the null; There is strong evidence against the null at
5% level of significance with equal variance assumed; It implies that there is significant difference
between average service quality for delivery and average base of the business. The claim is valid.

For Tangibles
H0: Average service quality for tangibles = Average base of business H1: Average service quality for
tangibles ≠ Average base of business

Two tail test Levene’s Test Pvalue=0.723 Level of significance= 5% (0.05)

Pvalue > Level of significance Equal variance assumed

Equality of Mean Test Two tail test Pvalue= 0.00 Level of Significance= 5% (0.05)

Conclusion Pvalue < Level of significance; We reject the null; There is strong evidence against the null at
5% level of significance with equal variance assumed; It implies that there is significant difference
between average service quality for tangibles and average base of the business. The claim is valid

RQ3: a) Goodness-of-fit for the type of businesses among the fish farm customers.

H0: The data “Type of Trading Business” among the fish farm customers is consistent with the specified
distribution H1: The data “Type of Trading Business” among the fish farm customers is not consistent
with the specified distribution Type: One tail test

Specified distribution: Retailers= 0.6 Wholesalers= 0.3 Restaurants= 0.1

Pvalue= 0.022 One tail test Level of significance= 0.05

Conclusion Since Pvalue < Level of significance; We reject the null; There is strong evidence against the
null at 5% level of significance; It implies that the data of the “Type of Trading Business” among the fish
farm customers is not consistent with the specified distribution. It doesn’t have uniform distribution OR;
The data does not have mentioned distribution

3 b) Association between the type of business and their base (Icelandic and non-Icelandic) {Test of
Association}

H0: Variable “Type of Business” & Variable “Base of Business” are independent. H1: Variable “Type of
Business” & Variable “Base of Business” are not independent.

Type: Two tail test Pvalue= 0.772 Level of significance = 0.05

Conclusion Since Pvalue > Level of significance; We do not reject the null; There is strong evidence for
the null at 5% level of significance; So the Variable “Type of Business” & Variable “Base of Business” are
independent; Thus, the type of business and base of business does not affect each other OR; the type of
business do not affect the base of business, The claim is valid

RQ4 b) Simple linear regression of the overall satisfaction & the best explanatory service quality
factor

Independent variable = Tangible service quality factor. Dependent variable = Overall satisfaction. (R-
square = 0.778/77.8%. The independent variable ‘tangible’ service quality factor is able to describe
77.8% variation in dependent variable ‘overall satisfaction’ in this model. Rest 22.2% (100%-77.8%) of
variation is described by other factors not accounted in this model.) Model check. H0: The model is not
good to show cause and effect relation. H1: The model is good to show cause and effect relation. P value
= 0.000. Level of sign = 0.05. (Regression model: overall satisfaction = b0 + b1 * tangible). The model
exists. For an independent variable Coefficient check. H0: The coefficient of tangible is not good for the
model. H1: The coefficient of tangible is good for the model. P value = 0.000. Level of sign = 0.05. P
value < Level of significance. Thus, the coefficient is good for the model. The model: Overall satisfaction
= b0 + b1 * Tangible is strong to show cause and effect relation. The coefficient of tangible is strong to
show cause and effect relation. The regression model is Overall satisfaction = b0 + b1 * tangible. Where,
b0 = 2.019 and b1 = 0.973. Overall satisfaction = 7.616 + 1.536 * Tangible

Interpretation

Irrespective of all independent variables, the minimum satisfaction one can get is 7.616. Note: When the
overall satisfaction of a person is 0, the value of tangible service is 7.616. Coefficient of tangible =
+1.536 (positive correlation). There’s positive correlation between overall satisfaction and tangible
service quality. Keeping everything constant when # of overall satisfaction increases by 1 time, the
tangible service quality also increases by 1.536 times. Note: Coefficient is positive here, so positive
correlation.

4c) Multiple regression of overall satisfaction and all three service quality factors considered
together.

Independent variable: Responsiveness, Tangibles and Delivery service quality factor. Dependent variable
= Overall satisfaction. Multiple linear regression model: Overall satisfaction = b0 + b1 * Responsiveness
+ b2 * Tangibles + b3 * Delivery. Model check: H0: The model is not good to show cause and effect
relation. H1: The model is good to show cause and effect relation. Coefficient check: For
responsiveness.H0: The coefficient of responsiveness is not good for the model. H1: The coefficient of
responsiveness is good for the model. For tangibles. H0: The coefficient of tangibles is not good for the
model. H1: The coefficient of tangibles is good for the model. For delivery. H0: The coefficient of
delivery is not good for the model. H1: The coefficient of delivery is good for the model. 2 nd model
should be used from spss output because the value of R – square in the 2 nd model is higher in comparison
to model 1 by 0.895/89.5%.

Interpretation of r-square: The independent variable ‘# of responsiveness, tangible and delivery’ service
quality factor together were able to describe 89.5% variation in dependent variable ‘overall satisfaction’
in this model. Rest 10.5% (100%-89.5%) of variation is described by other factors not accounted in this
model. Responsiveness service quality factor – Not retained. Tangible service quality factor – Yes
retained. Delivery service quality factor – Yes retained. Initial model: Overall satisfaction = b0 + b1 *
Responsiveness + b2 * Tangible + b3 * Delivery. Final model: Overall satisfaction = b0 + b2 * Tangible
+ b3 * Delivery (As responsiveness service quality factor was removed from the model). Overall
satisfaction = -5.370 + 1.443 * Tangible + 0.712 * Delivery.

Irrespective of all independent variables, the minimum overall satisfaction one can get is – 5.370.
Coefficient of tangible = 1.443. There’s positive correlation between tangible service quality factor and
overall satisfaction. Keeping everything constant, when tangible service quality factor increases by 1
time, the overall satisfaction also increases by 1.443. Coefficient of delivery = 0.712. There’s positive
correlation between delivery service quality factor and overall satisfaction. Keeping everything constant,
when delivery service quality factor increases by 1 time, the overall satisfaction also increases by 0.712.
We cannot establish cause and effect relationship between responsiveness service quality factor and
overall satisfaction.

4d) Multiple regression of overall satisfaction and all three service quality factors considered
together and the base of the trading business.

Independent variable: Responsiveness, Tangibles and Delivery service quality factor and base_dummy.
Dependent variable = Overall satisfaction. Multiple linear regression model: Overall satisfaction = b0 +
b1 * Responsiveness + b2 * Tangibles + b3 * Delivery + b4 * Base_dummy

Model check: H0: The model is not good to show cause and effect relation. H1: The model is good to
show cause and effect relation. Coefficient check: For responsiveness. H0: The coefficient of
responsiveness is not good for the model. H1: The coefficient of responsiveness is good for the model.
For tangibles. H0: The coefficient of tangibles is not good for the model. H1: The coefficient of tangibles
is good for the model. For delivery. H0: The coefficient of delivery is not good for the model. H1: The
coefficient of delivery is good for the model. For base_dummy. H0: The coefficient of base_dummy is
not good for the model. H1: The coefficient of base_dummy is good for the model. 2nd model should be
used from spss output because the value of R – square in the 3 rd model is higher in comparison to model 1
by 0.902/90.2%.

Interpretation of r-square. The independent variable ‘# of responsiveness, tangible, delivery and


base_dummy’ together were able to describe 90.2% variation in dependent variable ‘overall satisfaction’
in this model. Rest 9.8% (100%-90.2%) of variation is described by other factors not accounted in this
model. Responsiveness service quality factor – Not retained. Tangible service quality factor – Yes
retained. Delivery service quality factor – Yes retained. Base_dummy – Yes retained. Initial model:
Overall satisfaction = b0 + b1 * Responsiveness + b2 * Tangible + b3 * Delivery + b4 * Base_dummy.
Final model: Overall satisfaction = b0 + b2 * Tangible + b3 * Delivery + b4 * Base_dummy (As
responsiveness service quality factor was removed from the model). Overall satisfaction = -4.037 + 1.358
* Tangible + 0.584 * Delivery + 1.572 * Base_dummy.

Irrespective of all independent variables, the minimum overall satisfaction one can get is – 4.037.
Coefficient of tangible = 1.358. There’s positive correlation between tangible service quality factor and
overall satisfaction. Keeping everything constant, when tangible service quality factor increases by 1
time, the overall satisfaction also increases by 1.358. Coefficient of delivery = 0.584. There’s positive
correlation between delivery service quality factor and overall satisfaction. Keeping everything constant,
when delivery service quality factor increases by 1 time, the overall satisfaction also increases by 0.584.
Coefficient of base_dummy = 1.572. There’s positive correlation between base_dummy and overall
satisfaction. Keeping everything constant, when base_dummy increases by 1 time, the overall satisfaction
also increases by 1.572. We cannot establish cause and effect relationship between responsiveness service
quality factor and overall satisfaction.

You might also like