NY Times API Approach

Approach

I will use the New York Times Most Popular API and the 30-day viewed endpoint. My question is: Which New York Times sections appear most often among the most-viewed articles during the last 30 days?

The endpoint is https://api.nytimes.com/svc/mostpopular/v2/viewed/30.json. I will store my API key in .Renviron and read it with Sys.getenv("NYT_API_KEY") so the key is not written in the Quarto document. I will use httr to make the request and check the response status and content type. I will then use jsonlite to parse the JSON response.

I will convert the article records into a tidy data frame where each row represents one article. I will keep the fields needed for the question, including the article ID, publication date, section, title, byline, abstract, source, article type, and URL. I will convert the publication date into an R date, remove extra spaces, keep missing values visible, and remove duplicated article IDs.

Before doing the analysis, I will check the required fields, row counts, duplicate article IDs, missing titles, and blank sections. I will then count the number and percentage of articles in each section. The final table and chart will show which sections appear most often. I will not write the final finding until the live API results have been reviewed.

Sources

• The New York Times. API documentation. https://developer.nytimes.com/apis
• httr quickstart guide. CRAN. https://cran.r-project.org/web/packages/httr/vignettes/quickstart.html

AI Use

I used ChatGPT to help review the planned steps and organize the documentation.