library(tidyverse)## Warning: package 'tidyverse' was built under R version 4.0.3## -- Attaching packages --------------------------------------- tidyverse 1.3.0 --## v ggplot2 3.3.2 v purrr 0.3.4
## v tibble 3.0.4 v dplyr 1.0.2
## v tidyr 1.1.2 v stringr 1.4.0
## v readr 1.4.0 v forcats 0.5.0## Warning: package 'tibble' was built under R version 4.0.3## Warning: package 'readr' was built under R version 4.0.3## -- Conflicts ------------------------------------------ tidyverse_conflicts() --
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()library(tidytext)
library(knitr)## Warning: package 'knitr' was built under R version 4.0.3library(rvest)## Loading required package: xml2##
## Attaching package: 'rvest'## The following object is masked from 'package:purrr':
##
## pluck## The following object is masked from 'package:readr':
##
## guess_encodingCovid-19 has disrupted many industries worldwide including travel which has taken one of the biggest revenue losses out of all. With the hopes of an effective vaccine being distributed early next year, many believe that this milestone will bring back normalcy to our every day lives. Nonetheless, there will be a portion of the populaion that will be skeptical about the quick arrival of the vaccine and question its effectiveness of protecting their health and others’. It is difficult to say at this stage of the pandemic if the return to normal is on the horizon or not but the economy in a couple months will be one of the early indicators. Therefore, the truth that we will work to discover is if the Covid-19 vaccine will make the world feel comfortable to travel again.
# Read the data from the csv file
covid_data <- read.csv("Covid_CrowdSourcing.csv", header = TRUE)
kable(covid_data)# clean up the column names to remove periods
names(covid_data)[names(covid_data) == 'Article.URL'] <- "Article Url"
names(covid_data)[names(covid_data) == 'Length.of.Article'] <- "Length of Article"
names(covid_data)[names(covid_data) == 'Reliability.Coefficient'] <- "Reliability"
names(covid_data)## [1] "ID" "Category" "Article Url"
## [4] "Length of Article" "Weight" "Opinion"
## [7] "Reliability" "Sentiment"# remove the sentiment column. A brief sentiment analysis will be performed
# later in the code
covid_data <- select(covid_data, 1:7)
covid_data$`Article Text` <- ""
count = 1
article_text_col = which(colnames(covid_data) == 'Article Text')
for (val in covid_data$`Article Url`)
{
# Article URL
# Read the article data from HTML
# Only pull data from the body > div > p HTML tags
article <- read_html(val)
article <- article %>%
html_nodes("body") %>%
html_nodes("div") %>%
html_nodes("p") %>%
html_text()
cat("Article Text: ", val, "\n")
str <- toString(article)
covid_data[count, article_text_col] <- str
# keep track of how many articles are processed
count = count + 1
}## Article Text: https://www.pewresearch.org/science/2020/09/17/u-s-public-now-divided-over-whether-to-get-covid-19-vaccine/
## Article Text: https://www.bbc.com/news/health-54371559
## Article Text: https://www.bbc.com/news/health-54949799
## Article Text: https://www.traveloffpath.com/how-the-vaccine-will-impact-international-travel-in-2021/
## Article Text: https://www.bbc.com/news/entertainment-arts-54920146
## Article Text: https://www.cnbc.com/2020/11/09/travel-stocks-2nd-covid-waves-possible-impact-on-marriott-booking.html
## Article Text: https://www.honeywell.com/en-us/newsroom/pressreleases/2020/06/honeywell-to-introduce-fast-affordable-ultraviolet-cleaning-system-for-airplane-cabins
## Article Text: https://www.cntraveller.com/article/covid-vaccine-travel
## Article Text: https://www.cbsnews.com/news/how-unemployment-is-affecting-travel-bringing-families-closer/
## Article Text: https://www.wmur.com/article/how-could-a-covid-19-vaccine-could-change-travel-for-good/34676244#
## Article Text: https://www.travelagewest.com/Industry-Insight/Business-Features/Some-Say-a-COVID-19-Vaccine-Will-Be-a-Turning-Point-for-Travel-Will-It-?utm_source=NTGbox&utm_medium=web&utm_campaign=widgetkable(covid_data[1,])| ID | Category | Article Url | Length of Article | Weight | Opinion | Reliability | Article Text |
|---|---|---|---|---|---|---|---|
| 1 | 1 | https://www.pewresearch.org/science/2020/09/17/u-s-public-now-divided-over-whether-to-get-covid-19-vaccine/ | 6 | 2 | 8 | 0.9 | Numbers, Facts and Trends Shaping Your World, Science & Society, Science & Society, Pew Research Center conducted this study to understand how Americans are continuing to respond to the coronavirus outbreak. For this analysis, we surveyed 10,093 U.S. adults from Sept. 8 to 13, 2020. This report also draws on data from a survey fielded April 29 to May 5, 2020, among 10,957 U.S. adults., Everyone who took part in either survey is a member of Pew Research Center’s American Trends Panel (ATP), an online survey panel that is recruited through national, random sampling of residential addresses. This way nearly all U.S. adults have a chance of selection. The survey is weighted to be representative of the U.S. adult population by gender, race, ethnicity, partisan affiliation, education and other categories. Read more about the ATP’s methodology., See here to read more about the questions used for this report, along with responses, and its methodology., As efforts to develop and test a COVID-19 vaccine spur debate around the timing and release of a federally approved vaccine, the share of Americans who say they would get vaccinated for the coronavirus has declined sharply since earlier this year., About half of U.S. adults (51%) now say they would definitely or probably get a vaccine to prevent COVID-19 if it were available today; nearly as many (49%) say they definitely or probably would not get vaccinated at this time. Intent to get a COVID-19 vaccine has fallen from 72% in May, a 21 percentage point drop., The share who would definitely get a coronavirus vaccine now stands at just 21% – half the share that said this four months ago., There are widespread public concerns about aspects of the vaccine development process. On the heels of a pledge from nine pharmaceutical companies to ensure that a potential vaccine would meet rigorous standards, the Center survey finds three-quarters of Americans (77%) think it’s very or somewhat likely a COVID-19 vaccine will be approved in the United States before its safety and effectiveness are fully understood. And when asked about the pace of the vaccine approval process, 78% say their greater concern is that it will move too fast, without fully establishing safety and effectiveness, compared with just 20% who are more concerned approval will move too slowly, creating unnecessary delays., The new national survey by Pew Research Center, conducted Sept. 8-13 among 10,093 U.S. adults, finds intent to get a COVID-19 vaccine has declined across all major political and demographic groups., However, sizable differences across groups remain. Democrats and those who lean to the Democratic Party are 14 percentage points more likely than Republicans and Republican leaners to say they would probably or definitely get a vaccine (58% vs. 44%). And Black adults are much less likely to say they would get a vaccine than other Americans: Just 32% of Black adults say they would definitely or probably get a COVID-19 vaccine, compared with 52% of White adults, 56% of Hispanics and nearly three-quarters (72%) of Asian Americans. (Asian adults were interviewed in English only.), Concerns about side effects and uncertainty around the effectiveness of a vaccine are widely cited as reasons by those who would not get a COVID-19 vaccine if one were available today., Among the roughly half of Americans who say they would not get a COVID-19 vaccine, 76% say concern about side effects is a major reason why they would definitely or probably not get it., Several vaccines are currently under trial right now. One trial was temporarily put on hold earlier this month for potentially causing side effects in a trial participant, but has since resumed., A large majority (72%) of those who would not get a COVID-19 vaccine also say a desire to know more about how well it would work is a major reason why they don’t currently plan to get a coronavirus vaccine., Fewer adults cite not thinking they need the vaccine (31%) or the vaccine’s cost (13%) as a major reason they would not likely get vaccinated., Those who say they would definitely or probably get a vaccine for COVID-19 if it were available today see a range of factors that could impact that decision., Overall, 57% of those planning to get a vaccine say they would be a little (36%) or a lot (21%) less likely to do so if they had to pay out of pocket to get it. About four-in-ten (42%) say out-of-pocket costs would not change their likelihood of getting a vaccine., Similarly, majorities say that many people experiencing minor side effects (57%) and the vaccine being effective about 60% of the time (55%) would reduce the likelihood of them getting vaccinated at least a little. But fewer than two-in-ten say either of these things would make them a lot less likely to get the vaccine. The possible need to get a vaccine again every year or so is not seen as a major deterrent among those planning to get vaccinated: 70% say this wouldn’t make a difference to them., Researchers are still not sure how effective a COVID-19 vaccine will ultimately be. The U.S. Food and Drug Administration has said it would authorize a COVID-19 vaccine if it was safe and at least 50% effective in preventing the disease or decreasing the severity of infections, although Anthony Fauci, director of the National Institute of Allergy and Infectious Diseases, has said scientists are hoping for a vaccine that is at least 75% effective., As Americans look ahead to when a vaccine for COVID-19 is approved in the U.S., many express doubts about how safe and effective a vaccine will be initially., About three-quarters of Americans (77%) say it is at least somewhat likely that a vaccine for COVID-19 will be approved and used in the U.S. before it’s fully known whether it is safe and effective, including 36% who say this is very likely to happen. Just 22% say this is not too or not at all likely., Public assessments are more mixed when it comes to whether enough Americans will get vaccinated to curb the spread of the disease: 53% say this is at least somewhat likely, while 46% think it not too or not at all likely., Americans also have a mixed outlook on vaccine access. About half of U.S. adults (48%) say it’s at least somewhat likely that everyone who wants the vaccine will have quick and easy access to it, while 51% say this is not too or not at all likely., Consistent with the view that a vaccine may be approved before its safety and effectiveness are fully understood, Americans overwhelmingly say their greater concern is that the approval process will move too fast, rather than too slowly. Nearly eight-in-ten (78%) say their greater concern is that the vaccine approval process will move too fast, without fully establishing that it is safe and effective. Just 20% say they are more concerned the approval process will move too slowly, causing unnecessary delays in access to a vaccine., While Republicans and Democrats have differed over many aspects of the coronavirus outbreak – including the threat it presents to public health and how quickly to lift restrictions on public activity – majorities of both groups say their greater concern about the vaccine approval process is that it will move too fast, rather than too slowly. About seven-in-ten Republicans (69%) are more concerned about the approval process moving too fast, and an even larger majority of Democrats (86%) share this view., Those who plan to get a COVID-19 vaccine express much greater confidence in the vaccine development process than those who do not plan to get vaccinated., Overall, 19% of the public has a great deal of confidence that the research and development process in the U.S. will produce a safe and effective vaccine for COVID-19, while another 45% say they have a fair amount of confidence. About a third (35%) say they have not too much or no confidence in this process., Among those who say they would definitely or probably get a vaccine, more than eight-in-ten express either a great deal (30%) or a fair amount (54%) of confidence in the research and development process. By contrast, 55% of those not planning to get a coronavirus vaccine say they have not too much or no confidence at all in this process., , , 1615 L St. NW, Suite 800Washington, DC 20036USA (+1) 202-419-4300 | Main (+1) 202-857-8562 | Fax (+1) 202-419-4372 | Media Inquiries |
| , About Pew Research Center Pew Research Center is a nonpartisan fact tank that informs the public about the issues, attitudes and trends shaping the world. It conducts public opinion polling, demographic research, media content analysis and other empirical social science research. Pew Research Center does not take policy positions. It is a subsidiary of The Pew Charitable Trusts., To complete the subscription process, please click the link in the email we just sent you. |
There are various attributes being analyzed in order to perform an analysis and derive result based on the crowd sourcing data. The attributes being analyzed are the length (word count) of the article, the category weight, the opinion rating, the reliability rating, and the sentiment. Each attribute is explained further below:
The length of the article is measured by the actually word count of the article or text document. The images and image captions have been excluded from this value.
The category weight is the overall influence a particular category has on the outcome of the model. The weight is measured on a scale between 1 and 5, and the weight is assigned based on how many articles or data pieces the category contains. The more pieces of data in a category the higher the weight. A weight of 1 means a category has few data pieces, and a weight of 5 means the category has many data pieces.
The opinion rating represents how well the opinion reflected through the article aligns with our side of the argument. In other words does the opinion of the article support our side of the argument or not. The opinion is rated on a scale from 1 to 10, where 1 means it does not support our argument at all and 10 means it greatly supports our argument.
The reliability coefficient determines the reliability of the resource from which the data/article is coming, and determines the overall reliability of the data being presented in the article. This is rated on a scale from 0 to 1, where 1 is highly reliable and 0 is not reliable.
The sentiment determines if the content of the article aligns with our argument or not. Instead of reading it and making a opinionated decision like in the Opinion attribute, sentiment looks at the actual words used in the article and assigns a rating based on a scale from -2 to 2. -2 means the article did not support our argument while 2 means the article did support our argument.
The sentiment analysis being performed is not a full sentiment analysis. A bag of words was manually defined and two text files were produced. One text file contains key words that an article would use if it was providing data that supports the argument that the vaccine will not return work to normal. The other text file contains words that would support the opposite side of that argument. A value is produced between [-2,2], with -2 meaning it strongly does not support our argument and 2 meaning it strongly supports our argument.This is essentially a way of quickly analyzing through code whether or not a particular argument supports our argument. Using R allows the number of articles to be scaled without having to change the actual processing and R code.
# define undesirable words, These words will be excluded
# from the sentiment analysis
undesirable_words <- c("", "\t", "\n", "flexbox", "flex", "css",
"fax", "phone", "twitter", "instagram", "facebook",
"Address", "subscribe")
# Tokenize string into individual words, remove undesirables, remove stop words,
# remove words with fewer than 3 characters, remove words with more than 20 characters
covid_data_tidy <- covid_data %>%
unnest_tokens(word, `Article Text`) %>% # break lyrics into individual words
filter(!word %in% undesirable_words) %>% # remove all undesirable words
filter(!nchar(word) < 3) %>% # remove words less than 3 chars
filter(!nchar(word) > 20) %>% # remove words more than 30 characters
filter(!str_detect(word, ":")) %>% # remove strings with ':' (mainly CSS code)
anti_join(stop_words)## Joining, by = "word"# read not normal travel and normal travel words from csv file
normal_travel_words <- read.csv("Normal_travel_words.csv", header = TRUE)
# Get the sentiment of each article
covid_normal_travel <- covid_data_tidy %>%
group_by(covid_data_tidy$`Article Url`) %>%
inner_join(normal_travel_words) %>%
count(sentiment) %>%
spread(sentiment, n, fill = 0) %>%
mutate(sentiment = positive - negative)## Joining, by = "word"kable(covid_normal_travel)# Convert sentiment to a scale between -2 and 2
covid_scaled_sentiment <- covid_normal_travel %>%
mutate(sentiment_score = sin(sentiment)*2)
kable(covid_scaled_sentiment)# Add sentiment score to the original dataframe
covid_data$Sentiment <- covid_scaled_sentiment$sentiment_score
kable(covid_data[1,])| ID | Category | Article Url | Length of Article | Weight | Opinion | Reliability | Article Text | Sentiment |
|---|---|---|---|---|---|---|---|---|
| 1 | 1 | https://www.pewresearch.org/science/2020/09/17/u-s-public-now-divided-over-whether-to-get-covid-19-vaccine/ | 6 | 2 | 8 | 0.9 | Numbers, Facts and Trends Shaping Your World, Science & Society, Science & Society, Pew Research Center conducted this study to understand how Americans are continuing to respond to the coronavirus outbreak. For this analysis, we surveyed 10,093 U.S. adults from Sept. 8 to 13, 2020. This report also draws on data from a survey fielded April 29 to May 5, 2020, among 10,957 U.S. adults., Everyone who took part in either survey is a member of Pew Research Center’s American Trends Panel (ATP), an online survey panel that is recruited through national, random sampling of residential addresses. This way nearly all U.S. adults have a chance of selection. The survey is weighted to be representative of the U.S. adult population by gender, race, ethnicity, partisan affiliation, education and other categories. Read more about the ATP’s methodology., See here to read more about the questions used for this report, along with responses, and its methodology., As efforts to develop and test a COVID-19 vaccine spur debate around the timing and release of a federally approved vaccine, the share of Americans who say they would get vaccinated for the coronavirus has declined sharply since earlier this year., About half of U.S. adults (51%) now say they would definitely or probably get a vaccine to prevent COVID-19 if it were available today; nearly as many (49%) say they definitely or probably would not get vaccinated at this time. Intent to get a COVID-19 vaccine has fallen from 72% in May, a 21 percentage point drop., The share who would definitely get a coronavirus vaccine now stands at just 21% – half the share that said this four months ago., There are widespread public concerns about aspects of the vaccine development process. On the heels of a pledge from nine pharmaceutical companies to ensure that a potential vaccine would meet rigorous standards, the Center survey finds three-quarters of Americans (77%) think it’s very or somewhat likely a COVID-19 vaccine will be approved in the United States before its safety and effectiveness are fully understood. And when asked about the pace of the vaccine approval process, 78% say their greater concern is that it will move too fast, without fully establishing safety and effectiveness, compared with just 20% who are more concerned approval will move too slowly, creating unnecessary delays., The new national survey by Pew Research Center, conducted Sept. 8-13 among 10,093 U.S. adults, finds intent to get a COVID-19 vaccine has declined across all major political and demographic groups., However, sizable differences across groups remain. Democrats and those who lean to the Democratic Party are 14 percentage points more likely than Republicans and Republican leaners to say they would probably or definitely get a vaccine (58% vs. 44%). And Black adults are much less likely to say they would get a vaccine than other Americans: Just 32% of Black adults say they would definitely or probably get a COVID-19 vaccine, compared with 52% of White adults, 56% of Hispanics and nearly three-quarters (72%) of Asian Americans. (Asian adults were interviewed in English only.), Concerns about side effects and uncertainty around the effectiveness of a vaccine are widely cited as reasons by those who would not get a COVID-19 vaccine if one were available today., Among the roughly half of Americans who say they would not get a COVID-19 vaccine, 76% say concern about side effects is a major reason why they would definitely or probably not get it., Several vaccines are currently under trial right now. One trial was temporarily put on hold earlier this month for potentially causing side effects in a trial participant, but has since resumed., A large majority (72%) of those who would not get a COVID-19 vaccine also say a desire to know more about how well it would work is a major reason why they don’t currently plan to get a coronavirus vaccine., Fewer adults cite not thinking they need the vaccine (31%) or the vaccine’s cost (13%) as a major reason they would not likely get vaccinated., Those who say they would definitely or probably get a vaccine for COVID-19 if it were available today see a range of factors that could impact that decision., Overall, 57% of those planning to get a vaccine say they would be a little (36%) or a lot (21%) less likely to do so if they had to pay out of pocket to get it. About four-in-ten (42%) say out-of-pocket costs would not change their likelihood of getting a vaccine., Similarly, majorities say that many people experiencing minor side effects (57%) and the vaccine being effective about 60% of the time (55%) would reduce the likelihood of them getting vaccinated at least a little. But fewer than two-in-ten say either of these things would make them a lot less likely to get the vaccine. The possible need to get a vaccine again every year or so is not seen as a major deterrent among those planning to get vaccinated: 70% say this wouldn’t make a difference to them., Researchers are still not sure how effective a COVID-19 vaccine will ultimately be. The U.S. Food and Drug Administration has said it would authorize a COVID-19 vaccine if it was safe and at least 50% effective in preventing the disease or decreasing the severity of infections, although Anthony Fauci, director of the National Institute of Allergy and Infectious Diseases, has said scientists are hoping for a vaccine that is at least 75% effective., As Americans look ahead to when a vaccine for COVID-19 is approved in the U.S., many express doubts about how safe and effective a vaccine will be initially., About three-quarters of Americans (77%) say it is at least somewhat likely that a vaccine for COVID-19 will be approved and used in the U.S. before it’s fully known whether it is safe and effective, including 36% who say this is very likely to happen. Just 22% say this is not too or not at all likely., Public assessments are more mixed when it comes to whether enough Americans will get vaccinated to curb the spread of the disease: 53% say this is at least somewhat likely, while 46% think it not too or not at all likely., Americans also have a mixed outlook on vaccine access. About half of U.S. adults (48%) say it’s at least somewhat likely that everyone who wants the vaccine will have quick and easy access to it, while 51% say this is not too or not at all likely., Consistent with the view that a vaccine may be approved before its safety and effectiveness are fully understood, Americans overwhelmingly say their greater concern is that the approval process will move too fast, rather than too slowly. Nearly eight-in-ten (78%) say their greater concern is that the vaccine approval process will move too fast, without fully establishing that it is safe and effective. Just 20% say they are more concerned the approval process will move too slowly, causing unnecessary delays in access to a vaccine., While Republicans and Democrats have differed over many aspects of the coronavirus outbreak – including the threat it presents to public health and how quickly to lift restrictions on public activity – majorities of both groups say their greater concern about the vaccine approval process is that it will move too fast, rather than too slowly. About seven-in-ten Republicans (69%) are more concerned about the approval process moving too fast, and an even larger majority of Democrats (86%) share this view., Those who plan to get a COVID-19 vaccine express much greater confidence in the vaccine development process than those who do not plan to get vaccinated., Overall, 19% of the public has a great deal of confidence that the research and development process in the U.S. will produce a safe and effective vaccine for COVID-19, while another 45% say they have a fair amount of confidence. About a third (35%) say they have not too much or no confidence in this process., Among those who say they would definitely or probably get a vaccine, more than eight-in-ten express either a great deal (30%) or a fair amount (54%) of confidence in the research and development process. By contrast, 55% of those not planning to get a coronavirus vaccine say they have not too much or no confidence at all in this process., , , 1615 L St. NW, Suite 800Washington, DC 20036USA (+1) 202-419-4300 | Main (+1) 202-857-8562 | Fax (+1) 202-419-4372 | Media Inquiries | |
| , About Pew Research Center Pew Research Center is a nonpartisan fact tank that informs the public about the issues, attitudes and trends shaping the world. It conducts public opinion polling, demographic research, media content analysis and other empirical social science research. Pew Research Center does not take policy positions. It is a subsidiary of The Pew Charitable Trusts., To complete the subscription process, please click the link in the email we just sent you. | 1.818595 |
In order to make sure all the values going into the model are on the same scale, the values in the dataframe are normalized using a T-score. A T-score is used because the analysis is being run on a sample of the data available for this topic. Since we are using a sample size on the population we are using a T-score.
covid_data_t <- c("Length_of_Article", "Weight", "Opinion", "Reliability", "Sentiment")
covid_data_t$Length_Of_Article <- t.test(covid_data$`Length of Article`)## Warning in covid_data_t$Length_Of_Article <- t.test(covid_data$`Length of
## Article`): Coercing LHS to a listcovid_data_t$Weight <- t.test(covid_data$Weight)
covid_data_t$Opinion <- t.test(covid_data$Opinion)
covid_data_t$Reliability <- t.test(covid_data$Reliability)
covid_data_t$Sentiment <- t.test(covid_data$Sentiment)
# print t scores
covid_data_t$Length_Of_Article$estimate## mean of x
## 4.727273covid_data_t$Weight$estimate## mean of x
## 5.818182covid_data_t$Opinion$estimate## mean of x
## 6.909091covid_data_t$Reliability$estimate## mean of x
## 0.7272727covid_data_t$Sentiment$estimate## mean of x
## 0.2451984Now that we have all of the data needed and everything converted into a T-score, we can run the model. The model that was derived is the following: Model = d(x) + a(e)(b) a - weight of the articles x - length of the articles (word count) b - opinion d - reliability coefficient e - sentiment score Model - final output
The reliability and length of the article are multiplies together to correlate the reliability with the length of the article. Often times lengthier articles tend to be a more reliable source of information. The sentiment, weight, and opinion are multiplied together because how well the opinion aligns with our argument and the sentiment go hand in hand. Then to account for the number of articles in each category the weight is applied to this metric as well.
model_result <- (covid_data_t$Reliability$estimate*covid_data_t$Length_Of_Article$estimate) + (covid_data_t$Sentiment$estimate*covid_data_t$Opinion$estimate*covid_data_t$Weight$estimate)
cat("Model Result: ")## Model Result:model_result## mean of x
## 13.29459With a model calculation resulting in a 7.8. It is likely that travel will not go back to normal even if a vaccine is released for Covid 19. The words used for performing the sentiment were derived from scanning the articles and compiling a custom list of words. This results in some possible error in the sentiment analysis. It was also not a full sentiment analysis. A quick analysis was done to show the application. Then the articles are read into the code using web scraping. This can result from some error in the final text of the documents because the article text is being parsed from HTML tags.