Introduction
We are the first to map out the public sentiment concerning biotechnology. From now on, governments will increase investment in biotechnology. The increased investment may be in vaccine research or in public health or perhaps something more nefarious. The investment is also subject to political whims, which depend on public sentiment. In this article, we analyze the public sentiment behind biotechnology.
We divide the analysis in two parts: the first part uses google and the second part uses twitter. Modern life revolves around Google. To appreciate the pivotal nature of google, go back to April 2020 when an average person did not understand the COVID symptoms. According to a New York Times article titled Google Searches Can Help Us Find Emerging COVID-19 Outbreaks, author Seth Stephens-Davidowitz finds that Google searches can find COVID hotspots. For example, searches for ``I can’t smell" were the highest in New York, New Jersey, Louisiana, and Michigan. Infection rate was also the highest in those four states. Seth’s analysis uses Google Trends, which is a public tool that the research community can use to study anonymous and aggregate search data. In the same spirit as the article, we analyze relevant searches to understand the popularity of five words: biotechnology, bio-engineering, biotech, synthetic biology and syn-bio. Note that words like biotechnology may also be two words: bio and technology. Using google trends, we answer three key questions:
What are residents searching for when they google any of those keywords?
How does the interest in the keywords change through time?
How does biotechnology news differ with google queries?
In the second part, we analyze twitter responses. We analyze the key influencers and the associated sentiment tone.
Google Trends Analysis
Analysis of the five key words for web queries
To perform the analysis, we use the gtrendsR package in R. For the above mentioned keywords, we download weekly popularity index since 2004. For any keyword, Google Trends gives a normalized index of the relative use of the word. The trends index has a maximum of 100 where 100 represents the maximum search interest for the keyword. Also, Google Trends can measure search intensity in a region. In this way, by comparing several keywords, and the resulting queries, one can get a sense of a topic’s popularity. But the use of keywords is more of an art than an exact science. For example, a word like “biotechnology” may be two words: bio and technology. We chose the five keywords as a first start — there was no strategic direction.
What are residents searching for when they type any of those keywords?
Along with interest over time, for the keywords, we also download related queries. For example, for the keyword biotechnology, two associated queries are: What is biotechnology? and biotechnology jobs. For the five keywords, there were a total of 142 associated queries. To understand the relation between them, we separate each query into separate words. For example, the first query decomposes into three words: what, is and biotechnology. After the decomposition, we filter for stop words — we remove prepositions, articles, conjunctions and attachment words that do not convey any meaning by themselves. Then, the first query is left with only one relevant word: bio technology. The second query has two words" biotechnology and jobs.
Figure 1 shows the network of connected words left from the 142 queries. A few observations are in order. First, synthetic biology is associated with the word ACS which I assume is related to the academic journal. Therefore, synthetic biology may not be a popular term used by residents. Second, the word biotech is mostly associated with finance or economic terms. For example, searches for biotech involve startups, jobs, stocks, etf and etfs. In fact, upon inspection, a significant proportion of queries with biotechnology are also related to jobs, salary or careers. Last, the term engineering seems to be the most technical. The keyword engineering is used with bio, chemical, metabolic and genetic.
Table 1 shows the spatial dimension of the queries. For example, queries for biotech and biotechnology originated predominantly from South San Francisco. Bay area is also an active hub for biotech related jobs. The query for synthetic biology originated in the Boston Area which may be research related.
Figure 1 shows the network of words in queries related to the five keywords.
|
|
|
How does the interest in the keywords change through time?
Figure 2 shows the popularity of the five keywords over time. Note that the popularity is relative and the maximum popularity measure is indexed to 100. For clarity, we show five panels with each keyword highlighted in each of the panels. The results are stark. The words biotech and Biotechnology are more popular than words such as bio engineering, syn bio and synthetic biology. In fact, the popularity disparity is not even close. Relative to biotech and Biotechnology, the popularity of the other three words barely rises above zero.
Figure 2 shows the hits for the five different keywords over time.
How does biotechnology news differ with google queries?
Figure 3 compares the popularity from webqueries of biotech and Biotechnology keywords with interest garnered from news. Note that it is not appropriate to compare the index level of biotech with biotech news. From the bottom two panels, it is clear that the biotech and Biotechnology news follows a seasonal pattern. Also, the use of web queries for the keywords has decreased over time but the news with the keywords have increased.
Figure 3 shows the hits for biotech and Biotechnology compare with the hits generated from the news.