Quick Google Trends analysis
Alejandro Guizar
June 26, 2015
About Google Trends
Google Trends is a platform developed by Google Inc.to query how often a particular term is searched for in the Google search engine. The results are expressed in a normalized value (form 1 - 100), indicating the relative total search-volume from a given term within a given country or at the worldwide level (see https://www.google.com/trends).
The results returned by Google Trends are re-scaled by dividing the search-term hits obtained for a given week by the maximum number of hits obtained at any moment over the period of interest. As such, the returned values are not a measurement of the total number of searches of a particular term, but rather they provide a “popularity” indicator of a given term within a specific geographical context.
Results can be interpreted in the following way: a line trending downward means that a search term’s relative popularity is decreasing. It does’t necessarily mean the total number of searches for that term is decreasing (in fact, they could be increasing). It just means its popularity is decreasing compared to other searches. An important point to note here is that it is not possible to obtain the absolute value for the number of times a given term has been searched for.
Lastly, the results are also affected by the spatial extent considered in the analysis, thus the popularity trend for a term may vary if the search is focused on a single country or worldwide.
Quick trend analysis
Single-term searches
In this short exercise, I searched in Google Trend the following words, queried over the period 2004 - 2015:
- effectiveness
- standard
- label
- climate change
- adaptation
- socio-ecological
- landscape
The results plotted below indicate the relative popularity of these terms when searched individually. Only the word “socio-ecological” (and variations of the word, e.g. “socio-ecological”, “socio ecological” and “socioecological”) did not yield trend results, so this term was not included in the resulting plot.
##
## Attaching package: 'dplyr'
##
## The following objects are masked from 'package:stats':
##
## filter, lag
##
## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union
##
## Loading required package: GTrendsR## [1] "Substituting sparse Google Trends (monthly) with NA series"## No trend data for:
## "socio ecological"
Previous studies have shown that cyclic patterns in term-search results might be used as an indicator of biological processes (Proulux et al. 2014), and thus, web-crawling tools such as Google Trends could be used to track the timing of biological phenology with high precision (Dugas et al. 2012).
The time-series of some of the queried terms reveal a cyclic pattern that might be an indicator of such phenological processes, particularly the terms landscape and adaptation. It is not the intent of this report to throw any conclusions on possible correlations that derived from these cyclic patterns, but merely to denote the pattern shown by the data.
It is very important to note that Google Trends result need to be interpreted in the context in which a key word is used, as well as the overall context of the research carried out.
Combined term-searches
As a second phase of the research, I queried the Google trends database for the following combined terms:
- “effectiveness” and “standard”
- “effectiveness” and “label”
- “climate change” and “adaptation”
- “climate change” and “socio-ecological”
- “climate change” and “landscape”
- “adaptation” and “landscape”
This type of analysis include term-searches that contain the 2 terms in any order. For example, if we are interested in queries that contain the words ‘climate change’ and ‘adaptation’, the resting matches could include queries like “climate change adaptation”, “climate change and adaptation” or “adaptation to climate change”.
The plots below show the resulting trends of the combined search. No data was available for the combinations “effectiveness” and “label”, and climate change" and “socio-ecological”.
##
## Attaching package: 'zoo'
##
## The following objects are masked from 'package:base':
##
## as.Date, as.Date.numeric
References
Dugas, Andrea Freyer, Yu-Hsiang Hsieh, Scott R Levin, Jesse M Pines, Darren P Mareiniss, Amir Mohareb, Charlotte A Gaydos, Trish M Perl, and Richard E Rothman. 2012. “Google Flu Trends: Correlation with Emergency Department Influenza Rates and Crowding Metrics.” Clinical Infectious Diseases: An Official Publication of the Infectious Diseases Society of America 54 (4): 463–69.
Proulx, Raphaël, Philippe Massicotte, and Marc Pépino. 2014. “Googling Trends in Conservation Biology.” Conservation Biology: The Journal of the Society for Conservation Biology 28 (1): 44–51.