Quantitative Analysis of Usability Engineering with Galvanic Skin Response and Emotional Valence Response

This report presents a quantitative analysis of Usability Engineering with Galvanic Skin Response and Emotional Valence Response. The report includes data cleaning, data encoding, and missing data analysis. The report also includes descriptive statistics, quantitative analysis of Galvanic Skin Response, positive emotional valence response, and three research questions. The report concludes with outcomes and conclusions.

Running head: USABILITY ENGINEERING
Usability Engineering
Name of Student:
Name of University:
Course ID:

1USABILITY ENGINEERING
Table of Contents
Introduction:....................................................................................................................................2
Data Analysis:..................................................................................................................................2
1. Data cleaning, data encoding and missing data:......................................................................2
2. Descriptive Statistics:..............................................................................................................2
3. Quantitative Analysis of “Galvanic Skin Response”:.............................................................4
4. Quantitative Analysis of positive emotional valence response:..............................................6
5. Quantitative analysis of three research questions:...................................................................7
5.1. Research Question: “Does Arousal Response vary with different audio frequencies?”..7
5.2. Research Question: “Does perceived positive vividness response change with various
audio frequencies?”.................................................................................................................8
5.3. Research Question: Do the positive response of the three quantitative variables
“Speed”, “Strength” and “Straightness” equally respond as per the “High”, “Low” and
“Control” Audio Frequencies?................................................................................................9
Outcomes and Conclusions:..........................................................................................................10

2USABILITY ENGINEERING
Introduction:
The research report highlights the “Quantitative” data analysis collected for the analysis
of sector of “Usability Engineering” from the small survey of the participants. The considered
data analysis regards the experimental results with changing “Audio Interface” of the footwear
and examining the transformations in the behaviour of users. A total of 22 individuals took the
participation in the survey and the measures regarding experiment about footwear adjustment.
The objective of the study finds the information regarding shoe size, weight, height and physical
aspects for the gender. The individuals received the experimental responses of three different
types of frequencies “High Frequency”, “Low Frequency” and “Controlled Audio Frequency”
feedback while walking due to wearing the “Prototype Shoes”. MS-Excel software is utilised to
analyse the data.
For every aspects of this experiment, the analyst apprehended the viewpoints of the
individuals about their “Body Weight”, “Emotions”, “Mood”, “Physical Properties” and
“Behavioural Properties” in three dimensions.
Data Analysis:
1. Data cleaning, data encoding and missing data:
The responses of the participants are arranged in columns. A few missing values are
found in the data set. The considered data has numerous variables with missing values that
indicates some measures were not gathered. There are some columns also with missing data with
same types of participants. The total number of individuals with “Missing Data” is 6 out of 22.
The decision is to clean the data by removing 1 complete row. This row has missing value more
than 3. Removal of one single sample would not affect the study. The missing values are
replaced by the average values of the corresponding columns. In this way, the data is cleaned and
missing cells are either removed or filled. Data cleaning is the method of removing the
“Irrelevant” or “Missing” data. The analysis with missing values would make the analysis
clumsy and biased. Hence, data cleaning procedure is important and essential according to the
point of view of data analysis.
2. Descriptive Statistics:
The variable “Gender” is a categorical variable. Hence, its descriptive statistics is
declared by “Frequency Distribution”. The undertaken data includes 21 samples with 17 females
and 4 males that shows the females are greater in number than males.

3USABILITY ENGINEERING
The below analysis involves the descriptive statistics of numerical variables that are
“Age”, “Size of shoes”, “Weights in Kg” and “Heights in cm.” of the responders.
Descriptive statistics or summary statistics provides the capability to explore the
attributes of the data.
The mean age of the responders is 24.47 years with standard deviation 4.94 years. The
variable “Age” has a median 23. The maximum age of any person is 35 years and minimum age
of any person is 18 years with a range of 17 years. It is 95% probable that the mean “Age” lies in
the interval of 22.23 years to 26.73 years.
The average size of the shoes of the samples is 6.12 with the standard deviation of 1.82.
The median shoe size of the people is 6. The size of the shoes with occurs highest by size 5. The
size of the shoes varies in the interval of 10 to 3. The mean size of all the shoes lie in the range of
6.94 units to 5.29 units.
The mean “Weight” of the responders is 59.78 Kgs. with standard deviation 10.49 Kgs.
The median “Weight” is 57 Kgs. The highest weight of any sample is 89 Kgs. and least
“Weight” of any responder sample is 47 Kgs. The average “Weight” of all the samples between
64.56 Kgs. And 55.01 Kgs.

4USABILITY ENGINEERING
40 50 60 70 80 90 More
0
2
4
6
8
10
12
Weight (Kg)
Class
Frequency
The “Height” of the samples have average 165.14 cm and standard deviation of the
samples is 6.91 cm. The “Median” and “Mode” of the “Height” of the samples are both 165 cm.
The estimated mean lies in the interval of 162.00 cm and 168.28 cm.
150 155 160 165 170 175 180 More
0
1
2
3
4
5
6
7
Height (cm)
Class
Frequency
3. Quantitative Analysis of “Galvanic Skin Response”:

5USABILITY ENGINEERING
The table of “Galvanic Skin Response” shows that “Z-score” of “Galvanic Skin Response” is
higher for “High Audio Frequency” followed by “Moderate Audio Frequency” and least for
“Low Audio Frequency”.
 The mean “Z-score” for “High Audio Frequency” of “Galvanic Skin Response” ranges in
the interval of 0.04 to 0.21 with 95% possibility.
 The mean “Z-score” for “Low Audio Frequency” of “Galvanic Skin Response” ranges in
the interval of (-0.29) to 0.017 with 95% possibility.
 The mean “Z-score” for “Control Audio Frequency” of “Galvanic Skin Response” ranges
in the interval of (-0.31) to 0.26 with 95% possibility.
The bar plot of “Galvanic Skin Response” shows that average Z-scores of “Galvanic Skin
Response” for three types of audio frequencies are unequal.
Conclusion:
It could be concluded that “Z-scores” of “Galvanic Skin Response” vary with different
audio frequencies significantly.

6USABILITY ENGINEERING
4. Quantitative Analysis of positive emotional valence response:
The “Valence response” is positive when response rate is higher than 5. Among 21
observations, 17 samples have positive valance with “Higher Positive Frequencies”, 11
observations have “Low Positive Frequencies” and 13 observations have “Controlled Audio
Positive Frequencies”. The proportions of “Positive Valence Response” in “High Audio
Frequency”, “Low Audio Frequency” and “Control Audio Frequency” are found to be 0.81, 0.52
and 0.62 respectively.
 The estimated proportion of positive response in “High Audio Frequency” lies in the
interval of 0.64 to 0.98.
 The estimated proportion of positive response in “Low Audio Frequency” lies in the
interval of 0.31 to 0.74.
 The estimated proportion of positive response in “Control Audio Frequency” lies in the
interval of 0.41 to 0.83.
Conclusion:

7USABILITY ENGINEERING
It could be concluded that proportions of positive “Valence” vary with different audio
frequencies significantly.
5. Quantitative analysis of three research questions:
5.1. Research Question: “Does Arousal Response vary with different audio frequencies?”
The “Positive Arousal response” are the observations that have “Arousal” measure more
than 5. The proportion of positive “Arousal” frequency is higher for “High audio frequency”
followed by “Control Audio Frequency” and least for “Low Audio Frequency”.
 The estimated proportion of positive “Arousal” response with “High Audio Frequency”
varies from 0.21 to 0.64.
 The estimated proportion of positive “Arousal” response with “Low Audio Frequency”
varies from 0.056 to 0.42.
 The estimated proportion of positive “Arousal” response with “Control Audio
Frequency” varies from 0.09 to 0.48.

8USABILITY ENGINEERING
Arousal_High_Frequency Arousal_Low_Frequency Arousal_Control_Frequency
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
Arousal Response
Frequency Types
Positive Proportions
Conclusion:
It could be concluded that proportions of positive “Arousal” vary with different audio
frequencies significantly.
5.2. Research Question: “Does perceived positive vividness response change with various
audio frequencies?”
The “Positive Dominance” are the observations that have “Dominance” measure more
than 5. The proportion of positive “Dominance” is higher for “High Audio Frequency” and lesser
for both “High” and “Control” Audio Frequency simultaneously.
 The proportion of estimated positive “Dominance” with “High Audio Frequency” ranges
in the interval of proportion 0.41 and 0.83.

9USABILITY ENGINEERING
 The proportion of estimated positive “Dominance” with both “Low” and “Control”
Audio Frequency lies within the interval of proportion 0.21 to 0.64.
Conclusion:
It could be concluded that proportions of positive “Dominance” vary with different audio
frequencies significantly.
5.3. Research Question: Do the positive response of the three quantitative variables
“Speed”, “Strength” and “Straightness” equally respond as per the “High”, “Low” and
“Control” Audio Frequencies?

10USABILITY ENGINEERING
The positive proportions of the three variables “Speed”, “strength” and “Straightness” are the
variables whose response rate is more than 5. The positive proportional response rates refer that-
 The positive proportion of “Speed” response is greater for “High Audio Frequency”
followed by both “Control Audio Frequency” and “Low Audio Frequency”.
 The positive proportion of “Strength” response is greater for “High Audio Frequency”
followed by both “Control Audio Frequency” and least for “Low Audio Frequency”.
 The positive proportion of “Straightness” response is greater for “High Audio
Frequency” followed by both “Low Audio Frequency” and “Control Audio Frequency”.
Speed Strength Straightness
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Grouped bar plot of Postive proportions of three variables at three frequency
levels
High_Frequency Low_Frequency Control_Frequency
Variables
Positive Proportions
Conclusion:
Hence, it can be concluded that the positive proportions of the three considered variables
“Speed Response”, “Strength Response” and “Straightness Response” vary with different audio
frequencies.
Outcomes and Conclusions:
No statistical inferential test is applied in this analysis. The study finds how several
factors are influencing the fitness and preferences of “Footwear”. The research incorporates
several kinds of “Replicated” measures and its analytical understanding are interpretation for
“Decision-Making”. The 95% confidence intervals generated the “Upper” and “Lower”
confidence limits of the “Means” and “Proportions” of undertaken influencing factors. The
report mainly considers the usability and concepts of “Confidence Interval”. With the help of
“Visual observations" of “Confidence intervals” the conclusions about variability are drawn. In
future, the researcher should utilise various “Statistical inferences” for better results. Besides, all
the “Experimental factors” are not encompassed in the analysis. Henceforth, the scarcity of
considerate about “Footwear interface” and its consequences on choices may be generated.