Behavior-Change Goals for Social Networking Sites

HUCS Talk, March 20. Based on a CSCW paper and earlier work on Twitter regrets.

Transcript

1. I Would Like To…, I Shouldn’t…I Wish I…: Exploring Behavior- Change Goals for Social Networking Sites Patrick Gage Kelley patrickgagekelley.com @patrickgage pgk@unm.edu Manya Sleeper, Alessandro Acquisti, 
 Lorrie Faith Cranor, Sean A. Munson, 
 Norman Sadeh
3. Today I will focus on… Social Network Site (SNS) Privacy • Strong focus on theory/methodology • Metrics to measure privacy • Working with high schools in the state of New Mexico
4. The privacy “paradox”
5. “Americans say they are deeply concerned about privacy on the web and their cellphones. They say they do not trust Internet companies or the government to protect it. Yet they keep using the services and handing over their personal information. That paradox is captured in a new survey by Pew Research Center.” http://www.nytimes.com/2014/11/13/upshot/americans-say-they-want-privacy-but-act-as-if-they-dont.html
7. The privacy “paradox” “American adults are concerned about how the government and corporations are centrally collecting data about citizens and consumers, teenagers are freely giving up personal and private information in online journals. Marketers, school officials, government agencies, and online predators can collect data about young people through online teenage diaries. Herein lies the privacy paradox. Adults are concerned about invasion of privacy, while teens freely give up personal information. This occurs because often teens are not aware of the public nature of the Internet.” – Susan B. Barnes, First Monday, 2006
8. The privacy “paradox” explanations: People don’t care about privacy they say they do but they don’t really People don’t know what the impacts of sharing are they say they want privacy, but just because they think they should, not because they really understand the risks People can’t use the controls they really do want privacy, but can’t use the tools to find the balance of sharing/protection they want
9. Understanding privacy behaviors To really understand if there is a paradox… How do we measure privacy failures?
10. Understanding privacy behaviors Pruning and revising profile content is an important part of teens’ online identity management. 59% have deleted or edited something that they posted in the past. 53% have deleted comments from others on their profile or account. 45% have removed their name from photos that have been tagged to identify them. 31% have deleted or deactivated an entire profile or account. 19% have posted updates, comments, photos, or videos that they later regretted sharing. PewResearch, May 2013 Teens, Social Media, and Privacy
11. Regret as one measure of privacy Investigated this through a large 
 scale study of Twitter users
 What states of being lead to regret? What types of regret occurred? How did people become aware of regretted messages? What repair strategies did people use to cope with regretted messages? “I read my Twitter the next morning and was astonished” 
 A Conversational Perspective on Twitter Regrets. CHI 2013
12. Study design Two conditions: Conversational Twitter Asked to “recall an occasion when” they said or tweeted something and then regretted it Participants described: • Regret • Circumstances leading to regret • How became aware of regret • Repair strategies
13. Study design Conducted study on Amazon’s Mechanical Turk 3,175 participants started the survey, 946 disqualified
 (609 for posting less than once a month on Twitter) 1,879 participants, 601 could not recall regrets
 (456 for Twitter, 145 for conversation) Participants were paid $0.75 for completion Took 14.5 minutes on average
14. How do you ask about regret? Multiple different methods: The most regretted thing (life regrets) One regret probing for details (recency bias) Participants self report frequency (certain regrets might be more memorable, more frequently occurring, easier to report)
15. Types of regret Categories from Knapp, M. L., Stafford, L., and Daly, J. A. Regrettable messages: Things people wish they hadn’t said. Journal of Communication 36, 4 (1986), 40–58. Other Expressive Implied criticism Direct attack Blunder Reveal too much Direct criticism Percent of regrets 0% 6% 12% 18% 24% 30% Twitter Conversation
16. Twitter regrets target multiple people Types significantly more likely to be targeted at multiple people: Blunders (82%) Expressive content (84%) Content that revealed too much (80%) Including unintended audiences (13% for Twitter, 5% for conversational)
17. Repair strategies Percent  of  strategies 0% 10% 20% 30% 40% Percent  of  strategies 0% 10% 20% 30% 40% Delete Apologize Act like nothing happened Excuse Justify Say something to offset Deny Non-verbal behavior Other Twitter Conversation Percent of strategies
18. Twitter is different… Twitter participants tended to regret critical messages, often targeted at broad audiences Twitter participants became aware of regrets more slowly than conversational participants Twitter participants often relied on self-awareness or third- parties to tell them about regrets, absent physical audience cues Once aware of regrets, Twitter participants tended to delete the regretted tweet and/or apologize (though less than conversation)
19. So, if social networking sites are places for deleting embarrassing content, hiding drunk photos from employers, and generally putting one’s self into situations of regret, why are 1.23 billion people on Facebook… I Would Like To..., I Shouldn’t..., I Wish I...: Exploring Behavior-Change Goals for Social Networking Sites. Forthcoming at CSCW 2015
20. Understanding privacy behaviors another way… by using behavior-change goals to explore how people view SNSs as impacting their lives, and by describing the range of goals participants have.
22. Respondents had behavior change goals for the SNSs Not everyone had the same goals, and types of behavior-change goals vary by SNS Respondents described steps and facilitators for achieving goals
23. Finding a goal “the one thing you would most like to change about your behavior on” the SNS, which was referred to as their goal . Locke, E. A., and Latham, G. P. Building a practically useful theory of goal setting and task motivation: A 35-year odyssey. Am. psychologist 57, 9 (2002), 705.
24. We coded the free responses about the goals using five sets of codes • the goals themselves • reasons for the goals • benefits of the goals • steps to reach the goals • facilitators and barriers for the goals
25. Types of goals Post more (23%) Post better (17%) Use more (15%) Use better (13%) Use less (31%)
26. Types of goals FB Instagram Twitter 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Post more Post better Use better Use more Use less
27. FB Instagram Twitter 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Post more Post better Use better Use more Use less FB Instagram Twitter 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Post more Post better Use better Use more Use less Increase posting or site use FB respondents tended to want to post more to improve contact My  mom  is  also  on   Facebook  and  I  know   she  would  love  to   see  more  pictures  of   my  kids
28. FB Instagram Twitter 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Post more Post better Use better Use more Use less FB Instagram Twitter 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Post more Post better Use better Use more Use less Increase posting or site use Instagram and Twitter: tended to be for intrinsic reasons (happiness, better site use, responsibility) I  feel  like  I  am  not   using  Twitter  as  fully   as  I  can I  might  be  missing   important  information  by   skipping  over  certain   content
29. Reasons for goals Contact Time Intrinsic Attention Safety/security/privacy Self presentation Less bothering
30. Reasons for goals Attention Contact Intrinsic Less bothering Security/privacy Self presentation Time 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Post better Post more Use better Use more Use less
31. Goals participants didn’t have… Spend more time using game apps
32. Goals participants didn’t have… Spend more time using game apps Include more swear words in my posts Post more negative content Post less interesting content
33. Goals participants didn’t have… Spend more time using game apps Include more swear words in my posts Post more negative content Post less interesting content Post more content when I’m drunk Post more content when I’m upset
34. What does this tell us? People see SNSs as positively and negatively impacting their lives Varies by SNS Potential areas for behavior-change tools: • Productivity/time management • Increasing SNS use (e.g., use of specific features, making better use of site) • Improve content (e.g., for self-presentation, security/privacy)
35. Fogg Behavior Change Model BJ Fogg's Behavior Model www.behaviormodel.org/
36. ‹#› Making new/familiar actions permanent (21%): • self-monitoring • awareness • attitude changes • making it easier/ harder to perform a behavior Exercise  discipline Read  a  book  [instead   of  Facebook] Not  take  myself  too   seriously
37. ‹#› Improve or increase use by proactively creating better content/social connections Take  some  pictures  of   scenery Follow  more  people I  could  take  more   photos
38. ‹#› Use or post more through increasing/ decreasing cued behaviors Post  daily  on  Twitter I  could  make  an  effort  to   message  one  friend  a  week   on  Facebook  to  get  in  touch. Use less by decreasing cued behaviors
39. ‹#› Potentially helpful one-time actions: • Delete/install app • Adjust settings • Offline actions Take  a  socializing  class Research  companies  to   follow. I  could  delete  Facebook   from  my  bookmarks  to   get  it  out  of  my  face.  
40. Importance and difficulty Attention Contact Intrinsic Less bothering Security/privacy Self presentation Time 0% 25% 50% 75% 100% Agreement Disagreement or neutral 0% 25% 50% 75% 100% Importance Difficulty
41. Barriers and facilitators Participants also described current habits as a challenge (91, 17% overall) for reducing SNS use (43), but also for posting more (22) or better (13). Described SNS as an “addiction” and how the behavior change was not part of what they currently do, for example: “I’m too comfortable in my routine.” Technology-related factors were also common, such as SNS features or general ease of use (66), or easy site access or high levels of activity (53), primarily for goals like posting or using the site more.
42. Overall, goals varied by SNS, demonstrating that participants may associate different potential benefits and negative consequences with different SNSs Examining these goals, facilitators, and barriers allows us to explore potential mechanisms for helping users achieve desired behavior changes Designers of SNS-behavior-change interventions should also consider the broader context. SNSs have business models that rely on holding user attention. This may be consistent with only some behavior-change goals
43. Seen as the second most important type of goal Only 23 users reported a goal that we coded as security and privacy as a main reason Of these 21 of them wanted to post or use the sites better, only 2 wanted to use the sites less. Security and privacy in context
44. Instagram and Twitter participants tended to believe they should use the sites better, or would generally benefit from more posting or use, for example: “I feel like I am not using Twitter as fully as I can.” Some participants felt a sense of responsibility to increase use or posting: “Others follow me and I don’t write anything” or “I might be missing important information by skipping over certain content.” Intention of use
45. R E F E R E N C E S Kelley, P.G., Sleeper, M., and Cranshaw, J. Conducting Research on Twitter: A Call for Guidelines and Metrics. Measuring Networked Social Privacy Workshop at Computer-Supported Cooperative Work and Social Computing (CSCW) 2013. Sleeper, M., Cranshaw, J., Kelley, P.G., Ur, B., Acquisti, A., Cranor, L.F., Sadeh, N. “I read my Twitter the next morning and was astonished: A conversational perspective on Twitter regrets.” SIGCHI Conference on Human Factors in Computing Systems (CHI) 2013. Sleeper, M., Acquisti, A., Cranor, L.F., Kelley, P.G., Munson, S., Sadeh, N. “I would like to..., I shouldn’t..., I wish I...: Exploring behavior-change goals for social networking sites” Forthcoming at Computer-Supported Cooperative Work and Social Computing (CSCW) 2015 Suknot, A., Chavez, T., Rackley, N., Kelley, P.G. Immaculacy: A game of privacy. Student Game Design Contest. SIGCHI Annual Symposium on Computer-Human Interaction in Play (CHI PLAY) 2014. EAGER: Privacy's Sociocultural Divide across American Youth
 PI: Patrick Gage Kelley (Co-PI: Myra Washington) 
 National Science Foundation, 10/01/2014 – 09/30/2016, $242,832
46. T H A N K S T O Alessandro Acquisti, Tim Chavez, Lorrie Cranor, Justin Cranshaw, Steven Garcia, Matthew Kay, Michelle Mazurek, Sean Munson, Chris Ottino, Nathan Rackley, Norman Sadeh, Manya Sleeper, April Suknot, Myra Washington and all of my excellent colleagues and students at UNM Patrick Gage Kelley patrickgagekelley.com @patrickgage pgk@unm.edu