digital-divide/plot_survey.py at main · comnetsAD/digital-divide · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
import matplotlib.pyplot as plt
import matplotlib.gridspec as gridspec
import pandas as pd
import seaborn as sns
import textwrap
from scipy import stats

plt.rcParams['text.latex.preamble']=[r'\boldmath']
params = {
          'font.size' : 14,
          'legend.fontsize': 16,
          }
plt.rcParams.update(params)

plt.rcParams['ytick.labelsize'] = 16
plt.rcParams['ytick.labelsize'] = 16

plt.figure(figsize=(12,16))

gspec = gridspec.GridSpec(3, 5)

ax2 = plt.subplot(gspec[0, 0:2])
ax3 = plt.subplot(gspec[0, 2:])
ax4 = plt.subplot(gspec[1, 0:2])
ax5 = plt.subplot(gspec[1, 2:])
ax1 = plt.subplot(gspec[2, 0:])

df = pd.read_csv("data/survey_results_gilgit.csv", sep=";")

ques = 'Please indicate the extent to which you agree with the following statement: I occasionally avoid visiting certain websites because my Internet is too slow to load them.'

df2 = df[ques].value_counts(normalize=True)

sns.barplot(df2.index, df2.values*100, ax=ax1,\
            palette=['#d7191c','#fdae61','#ffffbf','#a6d96a','#1a9641'], hatch="//", edgecolor="black", \
            order=['Strongly agree','Somewhat agree','Neither agree nor disagree','Somewhat disagree','Strongly disagree'])

ax1.set_title("Please indicate the extent to which you agree with the following statement:\nI occasionally avoid visiting certain websites because my Internet is too slow to load them.")

max_width = 15
ax1.set_xticklabels(textwrap.fill(x.get_text(), max_width) for x in ax1.get_xticklabels())
ax1.set_ylim([0,50])
ax1.set_yticklabels(['0%','10%','20%','30%','40%','50%'])
ax1.set_ylabel("% of participants", fontsize=16)
ax1.text(-0.8,55, "c", fontsize=30, weight="bold")

##################################################
control = df[(df['_submitted_by']=="normal80")]
treatment = df[(df['_submitted_by']=="modified90")]
##################################################

ques = 'In terms of how the 4 websites looked, did you notice anything missing or out of the ordinary?'

c = control[ques].value_counts(normalize=True)
t = treatment[ques].value_counts(normalize=True)

x = ['Original Page','Lite-Web']
y = [c["Yes"]*100, t["Yes"]*100]

sns.barplot(x, y, alpha=1,palette=['#FAD933','#2b83ba'], edgecolor=["black"], ax=ax2)

#hypothesis test (fisher_exact)
c = control[ques].value_counts()
t = treatment[ques].value_counts()

oddsratio, pvalue = stats.fisher_exact([[c["No"], t["No"]],[c["Yes"], t["Yes"]]])

print ("oddsratio, pvalue:", oddsratio, pvalue)

ax2.plot([0, 0, 1, 1], [24, 33, 33, 29], lw=1, c="black")
ax2.text((0+1)*.5, 33, "ns", ha='center', va='bottom', color="black")
ax2.set_ylim(0,100)

ax2.set_ylabel("% of participants that answered \"Yes\"", fontsize=16)
ax2.set_yticklabels(['0%','20%','40%','60%','80%','100%'])
ax2.set_title('In terms of how the 4 websites\nlooked, did you notice anything\nmissing or out of the ordinary?')
ax2.text(-1,115, "a", fontsize=30, weight="bold")
###################################################

ques2 = 'Please rate the impact of the missing component(s) on the browsing experience'

t_yes = treatment[treatment[ques]=="Yes"]
t = t_yes[ques2].value_counts().reset_index()
t.columns = ['impact','count']
t['tc']='Lite-Web'

c_yes = control[control[ques]=="Yes"]
c = c_yes[ques2].value_counts().reset_index()
c.columns = ['impact','count']
c['tc']='Original Page'

sns.barplot(x='impact', y='count', data=c.append(t), alpha=1, ax=ax3,\
            palette=['#FAD933','#2b83ba'],edgecolor=["black"],\
            order=['High impact','Moderate impact','Slight impact','No impact'], hue='tc')

ax3.legend_.set_title(None)
ax3.set_ylabel("% of participants", fontsize=16)
ax3.set_xlabel("")
ax3.grid()

#calc the pvalue
def switch_to_num(argument):
    switcher = {
        "High impact": 3,
        "Moderate impact": 2,
        "Slight impact": 1,
        "No impact": 0
    }
    return switcher.get(argument, "Invalid entry")

c_yes['ans_num'] = c_yes[ques2].apply(lambda x: switch_to_num(x))
t_yes['ans_num'] = t_yes[ques2].apply(lambda x: switch_to_num(x))

# print(stats.ttest_ind(c_yes['ans_num'],t_yes['ans_num']))

max_width = 10
ax3.set_xticklabels(textwrap.fill(x.get_text(), max_width) for x in ax3.get_xticklabels())

ax3.set_title('If you chose yes, please rate the impact of the\nmissing component(s) on the browsing experience')
ax3.set_ylim(0,100)

######################################################
ques = 'In terms of how the 4 websited FUNCTIONED, did you notice anything missing or out of the ordinary?'

c = control[ques].value_counts(normalize=True)
t = treatment[ques].value_counts(normalize=True)

x = ['Original Page','Lite-Web']
y = [c["Yes"]*100, t["Yes"]*100]

sns.barplot(x, y, alpha=1,palette=['#FAD933','#2b83ba'], edgecolor=["black"], ax=ax4)

#hypothesis test (fisher_exact)
c = control[ques].value_counts()
t = treatment[ques].value_counts()

oddsratio, pvalue = stats.fisher_exact([[c["No"], t["No"]],[c["Yes"], t["Yes"]]])

print (oddsratio, pvalue)

ax4.plot([0, 0, 1, 1], [18, 22, 22, 20], lw=1, c="black")
ax4.text((0+1)*.5, 22, "ns", ha='center', va='bottom', color="black")
ax4.set_ylim(0,100)

ax4.set_ylabel("% of participants that answered \"Yes\"", fontsize=16)

ax4.set_yticklabels(['0%','20%','40%','60%','80%','100%'])
ax4.set_title(u'In terms of how the 4 websites\nfunctioned, did you notice anything\nmissing or out of the ordinary?')

ax4.text(-1,115, "b", fontsize=30, weight="bold")
#########################################################

ques2 = 'Please rate the impact of the missing component(s) on the browsing experience.1'

c_yes = control[control[ques]=="Yes"]
c = c_yes[ques2].value_counts().reset_index()
c.columns = ['impact','count']
c['tc']='Original Page'

t_yes = treatment[treatment[ques]=="Yes"]
t = t_yes[ques2].value_counts().reset_index()
t.columns = ['impact','count']
t['tc']='Lite-Web'


print (c['count'])
print (t['count'])

sns.barplot(x='impact', y='count', data=c.append(t), alpha=1,ax=ax5,\
            palette=['#FAD933','#2b83ba'], edgecolor=["black"],\
            order=['High impact','Moderate impact','Slight impact','No impact'], hue='tc')

ax5.legend_.set_title(None)

ax5.set_ylabel("% of participants", fontsize=16)
ax5.set_xlabel("")

# c_yes['ans_num'] = c_yes[ques2].apply(lambda x: switch_to_num(x))
# t_yes['ans_num'] = t_yes[ques2].apply(lambda x: switch_to_num(x))
# print(stats.ttest_ind(c_yes['ans_num'],t_yes['ans_num']))

max_width=10
ax5.set_xticklabels(textwrap.fill(x.get_text(), max_width) for x in ax5.get_xticklabels())

ax5.set_title('If you chose yes, please rate the impact of the\nmissing component(s) on the browsing experience')
ax5.legend()
ax5.set_ylim(0,100)
ax5.grid()

plt.tight_layout()
plt.savefig("figures/survey_results_gilgit.pdf", format="pdf")