Data Analysis Report on Bell Let’s Talk

Abstract

            In this paper, Bell Let’s Talk campaign has been analyzed. In order to analyze it, two main social media platforms were chosen. Using different search processes, data from both of the platforms were conducted. Both these platforms were chosen because the community was most active on them since these are used for interacting with users throughout the world. On the social media platform, Twitter, a specific approach was chosen for collecting the data using the data extractor tool or website, Netlytic. Facebook, another social media platform was used and a certain research process was chosen to get to the campaign. Using the same data extractor, information from the platform was collected. In order to analyze the information, Microsoft Excel was used.

Introduction

            Bell Let’s Talk is actually an advertising campaign which is designed by Bell Canada, a Canadian telecommunication organization for raising awareness while combatting the stigma that surrounds the mental health. Being an attempt towards CSR or corporate social responsibility to encourage the use of several platforms of social media just to engage people with the aim of raising awareness regarding the realities of mental health. It can be said that the most significant part is its name which is a social media campaign, carried out every year where money is donated by the corporation to the funds of mental health on the basis of prescribed interactions with their certain brand like the sheer number of messages, mainly SMS sent on their network along with the social media shares and posts with the hashtags. Therefore, this campaign is a positive influence on the mental awareness and it aims to spread information about mental health.

1.      Describe what search criteria you used to collect data from each of the two platforms and why.

[#BellLetsTalk (Day OR 2019 OR @BeIletstalk) Bell Let's Talk (@Bell_LetsTalk OR Bell Lets Talk]

            This is the criteria that I used for collecting data from the social media platform, Twitter. More precisely, I used the approach of ‘hashtags’ to find the campaign and the comments which were posted on it by the users. There are millions of tweets that were posted on the Bell Let’s Talk campaign and all the data was collected using the above mentioned approach. The approach was narrowed down by using the commonly term hashtags (Twitter, 2019).

            To collect data from the platform of Facebook, simply the URL of campaign was copied to the data extractor for gathering the information regarding posts (Facebook, 2019).

This way, data regarding Bell Let’s Talk Day was collected using Netlytic.

2.      Did you consider any other search criteria for Twitter and, if so, why did you decide against it?

There was another approach that could be used for collecting the data on the social media platform, Twitter and it was all about using simply the search using keywords rather than hashtags. This approach although could be used for searching the campaigns but it would miss the meaning of using Twitter. The approach of using hashtags was used because the campaign was originally used with the hashtag and it also assisted in getting to all the tweets that used the similar way of posting.

3.      Summarize your datasets in terms of their size and any other attributes that you feel are important.

Twitter:

Facebook:

Data set extracted from the Facebook page of Bell Lets talk is analyzed through the use of excel to find out the tendency of user contributed posts and system generated post (also known as bot- system post). The following table represents that the sum of all likes is only 5830 in the total data of 185 user’s posts. Same location and user ids present that these post are supported by the bot system or origination supporting bot posting for promotions. Somehow, analysis concluded that most of these posts are user contributed post that contains images, videos and status.   

4.      What research/analytical questions are you able to answer with your datasets? Provide 3–5 specific examples of possible questions? there are 3 sample questions presented below that can be answered through the use of these datasets:

                                i.            What is the tendency of likes by each user profile?

                              ii.            What type of posts are mostly system generated posts (videos or pictures)?

                            iii.            How many posts page is getting from the same location?

Indicate who would be interested in answering these questions and why (e.g., what organizations, professional roles, etc.).

In terms of Twitter, the question regarding which fields are created by users can be asked along with that of fields which are created by users in the context of Facebook. The last two questions concern the fields which are created by the both in terms of both Facebook and Twitter. These questions can professional roles (e.g., what organizations, professional roles, etc.???)? be asked by organization and the other professional representatives of the organizations (for instance in this case organization and professionals are IT support of the Bell Let talk page) to determine the understanding of the user.

In both of the platforms, author, description, title, and source are the fields which are created by users while other fields like guide and link are created by the bot.

5.      What are the major differences between the two datasets (Twitter and Facebook) in terms of the types of data that they contain? Hint: When you answer this question, keep in mind the types of data analysis that you can perform with the various data types.

One of the most prominent difference in both datasets is that metrics can be obtained using the data that is extracted from Facebook. This way, the analysis can tell about the success of the campaign that is present on Facebook. Furthermore, there is a broad range of data analysis that can be applied on the data that is obtained from Facebook and one such example is the predictive analysis.

6.      If you are asked to analyze the effectiveness of this social media campaign, what other data fields (that Netlytic did not retrieve) would you want to collect using the available APIs to help you with the analysis? Hint: Refer to the API documentation as discussed in class

 A specific field that I was unable to analyze what the time that the user spent on the official page of Bell Let’s Talk Day Campaign. This type of data would help in determining the interest of users and how they perceived the visuals which were displayed to an extent. Furthermore, the return-tweets data will help in responsiveness of the campaign.

Conclusion

            It can be concluded that Bell Let’s Talk Day campaign proved to be quite successful and in order to collect data about the posts, tweets, and messages, an extractor was used. To gather data, two platforms of social media were chosen namely, Facebook and Twitter. For both of the platforms, the approach was different and it gave data about the required campaign. For extracting the data, Netlytic, a data extractor platform was used. The approach of hashtags was used since it is a trademark approach for the campaign. For producing further report, both datasets were compared and analyzed. There is an opportunity for improvements in the APIs which are available for collecting the data.

References

Facebook. (2019, Jan 30). Bell Let's Talk. Retrieved from Facebook: https://web.facebook.com/BellLetsTalk/?_rdc=1&_rdr

Twitter. (2019, Jan 30). Bell Let's Talk. Retrieved from Twitter: https://twitter.com/Bell_LetsTalk?ref_src=twsrc%5Egoogle%7Ctwcamp%5Eserp%7Ctwgr%5Eauthor