Using APIs

SICSS, 2022

Christopher Barrie

Introduction

• Why count tweets?

  • Test your API call

  • Of substantive interest

    • But… design considerations (normalizing)

Counting tweets

• As barometer of public opinion

)

Source: here

Counting tweets

• As measure of issue salience

Source: here

Counting tweets

• As archival source

Source: here and here

Design considerations

• Is this a barometer of offline opinion…?

Design considerations

• Is this a barometer of offline opinion…?

Source: here

Design considerations

• Is this an appropriate measure of issue salience?

  • What % of MPs are on Twitter?

Design considerations

• Is this an appropriate measure of issue salience?

  • What % of MPs are on Twitter?
  • Answer: ~90% (see this useful website)

    • Coverage questions similar to survey design:

      • Does my sample capture population of interest?

        • What is the coverage?

        • Is there temporal bias?

        • Is there (equivalent) non-response?

Design considerations

• Is this an appropriate measure of issue salience?

  • Does this capture online and offline dimensions?

    • Is communication here domain-specific?

    • How could we tell?

Group work

1. Come up with a question answerable using get_tweet_counts()
2. Formulate code needed to request that data
3. (If access granted) make the API call to get data
4. Consider limitations + design considerations

   • What bias might be built in?

   • Does this illuminate dynamics outside online domain?

   • What is the coverage like?