Scholars of technology often employ a social constructivist approach, and presume that social groups appropriate technologies and use them to advance their own ends (Song 2009; Baym 2010).  Social constructivists emphasize that "social structural agents and resources play highly formative roles in the production, implementation, and use of a technology" (Song 2009, p. 20-21).  Several studies use this approach to illustrate how groups in various contexts have used information and communication technologies (ICTs) for their own group specific needs (Byrne 2008; Campbell-Grossman et al. 2009; Stern & Adams 2010). The digital practice perspective extends this social constructivist framework, while focusing on the consumption of technology (as opposed to the socially constructed nature of invention or distribution of technology).  A digital practice perspective also extends the digital divide and digital inequalities frameworks. Digital practice suggests that individuals use the internet for practical purposes in order to accomplish daily activities. Contrary to digital divide/inequality approaches, there is an understanding that more education or income does not necessarily lead to more access or to different online usage.  Instead, the particular needs and objectives of particular groups contribute to greater or lesser ICT usage.

    Thus, the digital practice perspective is based on the understanding that social groups often appropriate technologies and use them to advance their own ends (Warschauer 2004; Song 2009).  Warschauer (2004, p. 206) suggests using a sociotechnical model to understand the dynamics of technology diffusion in which scholars "look at what people do rather than merely at what equipment they have." A premise of this perspective is that technology that is introduced into society does not determine people's actions, as much as people's particular experiences and objectives determine how a technology is or is not used.  Digital practice does not focus on whether or not a group is using a technology in the manner it was intended (i.e. focusing on the degree to which a group is using Facebook for its intended purpose of developing and maintaining social ties). Rather, the framework focuses on how a group has appropriated a technology to meet a context specific goal(s) (i.e. focusing on how a group uses the capabilities of Facebook to reach desired ends, of which it may or may not be to develop and maintain social ties).  Mehra et al. (2004) discuss several studies in which previously disadvantaged groups became internet users by incorporating the capabilities of the internet into their daily lives.  Similarly, Campbell-Grossman et al. (2009) show how low income African American mothers, after being given internet access, were able to effectively create a social support system by using e-mail. 

    Employing a digital practice perspective means paying less attention to models of technological deprivation.  Instead, digital practice reorients analysis of ICT activity towards understanding how, and more importantly why we observe particular patterns of ICT usage.  The focus on revealing unique patterns of ICT usage and understanding why these patterns are present deviates from standard approaches of domestication of ICTs.  For example, Kraut et al.'s (2006) edited volume compiles various studies of ICT usage in everyday life.  Their focus is less on why people choose certain activities online, and instead is on the social impact of these choices. An important tenet of this approach is that in exploring disparities in ICT activity, it does not assume that the activities of traditionally advantaged groups represent the default standard.  Further, Hargittai and Walejko (2008) show that students with higher socioeconomic statuses are more likely to produce online content as compared with students with lower socioeconomic status. Their finding can be interpreted as indicating a skills divide (van Dijk 2005). However, the benefit of producing online content, for example, having your groups' culture on the internet, may not be paramount for college students from lower socioeconomic backgrounds.  While the disparities in producing content are real, it is possible that students not producing online content are investing their time online with other activities, and presumably reaping disproportionate benefits in those areas.  The digital practice perspective would turn attention to these areas.   
 
    The argument for the digital practice approach is based on the ubiquity and user-friendliness of ICTs, such that measurable differences in resources are not always accurate predictors of ICT activity. This paper presents the digital practice approach to understanding differences in ICT usage and the benefits accrued through this usage.  I argue that this approach is most appropriate for societies in which access is no longer the primary issue, and in which ICTs have become domesticated (Haddon 2006).  I explain this approach and present an application of digital practice through the use of social networking sites (SNS) by African Americans.

    Two trends provide support for the digital practice perspective. First, technology has become less cost-prohibitive over the past several decades.  Second, publically available hardware and software have made usage of ICTs feasible for even the most technologically challenged groups.  For example, the idea of surfing the web may become antiquated as search engines (e.g. Google, Bing), web portals (e.g. Yahoo, AOL), and social networking sites (e.g. Facebook, Blackplanet) provide news and information directly to the user without that user having to leave the site (Pariser 2011).  Also, many companies are now tethering the hardware they sell to consumers (Zittrain 2008).  This practice allows these companies to provide licensed software applications to their users and make periodic updates to the software.  To illustrate, someone who purchases a blue-ray player, television, or e-reader, will be able to use these software applications without knowledge of how to locate, download, and operate these applications.  These trends have led to the "taming of wild" technologies (Baym 2010) such that software applications including social networking sites and e-mail and hardware such as cell phones, laptops, and tablet computers are increasingly integrated into everyday life.  While technologically sophisticated groups can modify their computing environment to the point of writing codes or manipulating hardware, the majority of benefits that can be gained from ICTs are attainable with little expertise.


    The digital practice perspective is based on the knowledge that impediments to accruing the benefits of ICT usage have been reduced significantly in recent years.  Groups can now use ICTs as they deem necessary to advance their own objectives.  Internet scholars have amassed a wealth of evidence which shows that the internet has allowed people to cultivate social and professional ties and to meet diverse other objectives online, through virtual communities. These observations lead to two research questions about how African Americans use social networking sites: 

1.  Are African Americans more likely to join social networking sites than are other ethnoracial groups? 
2. Are African Americans more likely to engage in social networking activities than are other ethnoracial groups?
   

    I use nationally representative data to examine differences between African Americans and other ethnoracial groups regarding their use of social networking sites.  The data for this research comes from the Pew Internet and American Life's "Spring Tracking Survey 2008". The Pew Internet and American Life Project conducts original research that explores the impact of the internet on individuals, families, communities, and on social, economic, political and religious institutions. The survey was conducted between April 8th, 2008 and May 8th, 2008 through telephone interviews of a random sample of adults age 18 or over.  The survey asks questions about respondents' internet activities and attitudes towards technology.  The response rate is 25%.  The total number of respondents is 2251, with African American respondents numbering 215. Pew provides a weight for this survey derived from the Census Bureau's March 2007 Annual Social and Economic Supplement to correct for response bias. A description of the sample is provided in Table 1.

Independent Variables

    The independent variables for the analyses represent standard demographic and socioeconomic predictors and are presented in Table 1.  These variables are education (less than high school, high school, some college, and college graduate), income, gender (male and female), community status (urban, suburban, and rural), marital status (married, cohabitating, divorced-separated-widowed, never been married or single), and ethnoracial status (white, black, Hispanic, other).  The ethnoracial variable is of primary importance. 

Variable	Freq (%)	Variable	Freq (%)
Education (N=2220)		Community Status (N=2251)
Less than High School	8.9	Urban	26.5
High School	32.3	Suburban	50.7
Some College	26.0	Rural	22.7
College Graduate	32.7
		Marital Status (N=2217)
Income (N=1755)		Married	55.7
Less than $20,000	17.2	Cohabiting (Living with partner)	3.6
Between $20,000 and $39,999	24.2	Divorced, Separated, or Widowed	27.6
Between $40,000 and $100,000	41.2	Never Been Married	11.8
Over $100,000	17.5	Single	1.4
Gender		Ethnoracial Status (N=2191)
Male	45.5	White	81.8
Female	54.5	Black	9.8
		Hispanic	4.8
		Other	3.6
Continuous Variables
	Mean	SD
Age	55.1	19.48

Dependent Variables

    Two sets of questions were predicted.  First, general social networking activity was assessed as how many sites respondents have profiles on.  Second, engagement with activities that foster the development of digital practices was assessed. These activities are described as affective (more personal or emotional) and instrumental (more professional).  Affective activities were measured by the survey questions: (1) Do you stay in touch with friends, and (2) do you make plans with your friends?  Instrumental activities were measured by the survey questions: (1) Do you make new business or professional contacts, (2) do you promote yourself or your work, and (3) do you make new friends?  Appendix A lists the survey questions and the summary statistics for these questions. 


    Binary logistic regression models were run to determine the effect of being African American on social networking activities.  Logistic regression was used since the available measures are binary.  One measure is ordinal (how many sites does the respondent have profiles on), but given the bi-modal distribution of the variable, it was recoded as binary.  For each variable, the odds ratio are presented of being in one category of the dependent variable (coded 1), as opposed to being in the other category (coded 0).  All dependent variables are presented as "yes" or "no" questions, such that 1 = yes, and 0 = no.  An odds ratio greater than 1 for an independent variable indicates that the variable's effect is to increase the odds of "yes".  Conversely, an odds ratio of less than 1 indicates that the variable's effect is to decrease the odds of "yes".  Values above 1 represent percentage point increases, and values below 1 represent percentage point decreases.  Similar to parameter estimates for ordinary least squares regression analysis, an odds ratio for nominal variables indicate an increase in odds compared to the reference variable.  For example, if in the model predicting having multiple social networking profiles we observe an odds ratio of .459 for income between $20,000 and $40,000, the odds can be interpreted as: "The odds of having profiles on multiple social networks for respondents with incomes between $20,000 and $40,000 is 54% less than for respondents with incomes between $40,000 and $100,000."

    Demographic and economic predictors for each model are included.  The main purpose of their inclusion is to test the effect of being African American net of these predictors.  Results that are not significant are not discussed.   


General Social Networking Activity

    The odds ratios for activities performed on general social networking sites are presented in Table 2.  Older ages, having a high school diploma, being married, and living in rural areas decrease the odds of being an SNS user.  Being Hispanic and of another ethnoracial group increases the likelihood of being an SNS user.  Older ages, higher incomes, being female, having less than a high school diploma, and being single decrease the odds of being a multiple site user, whereas having lower income and being divorced, separated, or widowed increase these odds.

    In examining the effect of being African American, the results show that African Americans are not more likely than whites to be SNS users.  However, African Americans are over twice as likely (2.57 times) as are whites to have multiple SNS profiles. This result was quite large, and since the focus is on African Americans, being a multiple SNS user was predicted again, this time with African American as the reference variable.  This way, African Americans can be compared with other ethnoracial groups.  In this model, African Americans are much more likely than all other racial groups measured to have profiles on multiple SNS.  Specifically, whites are 53% less likely, Hispanics are 51% less likely, and respondents of other ethnoracial groups are 41% less likely to have profiles on multiple SNS. 

	SNS User	Multiple SNS User?	Multiple SNS User (African American as Reference
Variable	B	B	Variable	B
Age	.929***	.983**	Age	.982**
Female	.961	.509**	Female	.514***
Income (Between $40,000 and $100,000 Reference)			Income (Between ($40,000 and $100,000 Reference)
Less than $20,000	2.238***	1.447*	Less Than $20,000	1.436*
Between $20,000 and $40,000	.974	.459***	Between $20,000 and $40,000	.445***
Over $100,000	.944	.452***	Over $100,000	.493***
Education (High School Reference)			Education (High School Reference)
Less than High School	1.638**	.502**	Less than High School	.502**
Some College	1.666***	.945	Some College	.922
College Graduate	1.889***	1.180	College Graduate	1.169
Community Status (Urban Reference)			Community Status (Urban Reference)
Suburban	.914	1.100	Suburban`	1.109
Rural	.689**	1.125	Rural	1.106
Marital Status (Married Reference)			Marital Status (Married Reference)
Cohabiting	3.155***	1.412	Cohabiting	1.442
Divorced, Separated or Widowed	1.787***	2.373***	Divorced Separated or Widowed	2.407***
Never Been Married	2.054***	1.059	Never Been Married	1.056
Single	1.231	.325*	Single	.359*
Ethnoracial Status (White Reference)			Ethnoracial Status (Black Reference)
Black	1.023	2.568***	White	.469***
Hispanic	1.441	1.056	Hispanic	.487**
Other	1.626**	1.271	Other	.586
Constant	3.496***	1.707	Constant	3.846***
Nagelkerke Rsquare	.348	.133	Nagelkerke Rsquare	3.846***

*p<=.05, **p<=.01, ***p<=.001


Activities which promote digital practices

    Although African Americans are not more likely to be SNS users as compared to whites, they are more likely than all ethnoracial groups to join multiple SNS. The next focus is on the specific activities completed by African Americans on SNS.  Since African Americans had higher odds than all other ethnoracial groups of being multiple SNS users, being African American was treated as the reference category for additional models.  In Table 3, five activities were measured. These activities are categorized as indicators of digital practices.

Affective digital practices
Activity 1 - Stay in touch with friends?
Activity 2 - Make plans with your friends?

Instrumental digital practices
Activity 3 - Make new business or professional contacts?
Activity 4 - Promote yourself or your work?
Activity 5 - Make new friends?

	Affective Model 1	Affective Model 2	Instru- mental Model 3	Instru- mental Model 4	Instru- mental Model 5
Variable	Odds-Ratio	Odds-Ratio	Odds-Ratio	Odds-Ratio	Odds-Ratio
Age	.909***	.912***	.948***	.947***	.928***
Female	1.222**	1.161	.491***	.497***	.890
Income (Between  $40,000 and $100,000 Reference)
Less than $20,000	1.453***	1.194	1.105	.922	1.596***
Between $20,000 and $40,000	1.050	.662***	.845	.866	.736**
Over $100,000	.906	.947	1.0601	.974	.506***
Education (High School Reference)
Less than High School	1.367*	1.208	1.120	2.320***	1.302
Some College	2.477***	2.422***	1.995***	1.771***	1.370***
College Graduate	2.235***	1.000	3.027***	2.546***	1.328*
Community Status (Urban Reference)
Suburban	.939	.741**	.606***	.786*	.675***
Rural	.778*	.725*	.565**	.523**	.608***
Marital Status (Married Reference
Cohabiting	2.195***	2.478***	2.660***	1.859**	1.967***
Divorced, Separated or Widowed	1.543***	1.786***	.965	1.008	2.228***
Never Been Married	1.772***	1.955***	1.003	1.335*	2.107***
Single	.960	.531	.908	1.230	1.577
Ethnoracial Group (Black Reference)
White	1.307*	1.134	.474***	.524***	1.192
Hispanic	1.578**	1.199	.763	.862	1.402*
Other	1.744**	.650*	.359***	1.000	.869
Constant	3.306***	2.707***	9.84	.769	1.511
Nagelkerke Rsquare	.403	.363	.170	.185	.284

*p<=.05, **p<=.01, ***p<=.001    

   Socioeconomic variables are important for both affective and instrumental practices.  As age increases, the odds of doing any of the activities are significantly reduced.  Users with educational attainment higher than a high school diploma have higher odds of doing any activity.  This is also true for users with less than a high school education.  Being married is associated with lower odds of doing any activity.  The analyses present interesting findings as to how being African American impacts affective and instrumental practices.
 
    For affective activities, African Americans do not use SNS significantly more than do other ethnoracial groups. Being African American lowers the odds of staying in touch with friends (activity 1). Whites are 1.3 times more likely, Hispanics are 1.58 times more likely, and respondents from other ethnoracial groups are 1.74 times more likely to use social networking sites to stay in touch.  Regarding making plans with friends, other ethnoracial groups are 35% less likely than African Americans of doing this activity, but there are no differences between African Americans and whites or Hispanics.

    African Americans have greater odds of making new business and professional contacts using SNS than all groups except Hispanics (activity 3). African Americans are also more likely to promote themselves or their work on SNS than are whites (activity 4).  Similar to model 3, there are no significant differences between African Americans and Hispanics with regard to promoting yourself or your work.  Finally, for activity 5, making new friends, African Americans are not significantly different from whites, and are less likely than Hispanics to use SNS for making new friends. Being Hispanic increases the odds of using SNS to make new friends.
 
    The research questions can be evaluated based on these findings. In response to the first question (are African Americans more likely to join social networking sites than are other ethnoracial groups?), the findings strongly show that African Americans are more likely to join these online sites than are white Americans.  Although there is no significant difference between African Americans and whites in having a profile, African Americans are more than twice as likely to have profiles on more than one SNS.  The second research question posed whether or not African Americans would more likely engage in social networking activities than would other ethnoracial groups.  The results show that African Americans are indeed more likely to make professional contacts and promote themselves or their work on SNS, as compared with other ethnoracial groups. Given affirmation for the research questions, we can say that African Americans do engage in SNS activities to achieve instrumental professional objectives.   

    The results show that Hispanics were more likely to use SNS than other groups, and that this use was motivated by making friends, irrespective of affective and instrumental activities.  Hispanics were more likely than African Americans to make new friends and to stay in touch with friends. This finding also provides affirmation for the research questions. If, similar to African Americans, Hispanics receive less utility from their professional offline social networks, then similar to African Americans, Hispanics would be expected to use the internet to mitigate this disparity. The analysis is limited in that the data was insufficient for other sizable minority groups such as Asians. 

    This study applies the digital practice approach to examine how African Americans use social networking sites. The aim of the research is to provide further understanding of differences in ICT usage among different groups.  Based on previous findings that significant benefits accrue from ICT usage for the majority of people, a rationale was presented for eschewing the notion of technological deprivation.  I argue instead for a reorienting of analysis of ICT activity towards understanding how social groups have appropriated technologies to meet their own needs.  In order to illustrate this perspective, I examined the social networking activities of African Americans, and found support for the argument that a "digital practice" among African Americans is the development of instrumental activities through social networking.  Differences in skill sets, economic resources, and motivation that produce significant differences between groups in society do exist.  Yet these factors and the differences they produce must be placed within a broader structural context. 

    Prevailing explanations of the digital divide emphasize disparities based on socio-economic and demographic characteristics. These explanations are partially true, as people most affected by these disparities also show increasing access to and use of technologies. Most research on ICT access and usage has focused on disadvantages between different groups. This study develops a framework for systematic examination of ICT usage differences within a group, which situates the digital divide and digital inequalities model within a broader conceptual model of digital practice, exemplified by how groups of people use ICTs. While not the first to document ways in which digital practices influence ICT usage (see Kraut et al. 2006), this study presents a distinctive framework that has not been fully examined.

    The digital practice framework brings the goals and objectives of specific groups to the forefront in re-conceptualizing how people use ICTs. This framework provides the opportunity for broader theoretical insights into how people use ICTs based on the context of their social constraints and their instrumental needs. The argument is presented that as more people have access to and use ICTs, digital practices become more salient in explaining their use. Affective and instrumental activities are used to explore how digital practices emerge among African Americans, and to explore some consequences of these practices.

    This study contributes to the literature by showing the relevance of the digital practice framework for examination of ICT use, and how this framework can change the ways in which we understand social inequalities and how they impact different social groups. The increased likelihood of African Americans joining SNS can be situated within the context of historical and contemporary acts of exclusion. It is clear that African Americans are at a social and economic disadvantage in the United States.  This disadvantage has been manifested within the digital divide or digital inequality frameworks. If African Americans are disadvantaged based on multiple socioeconomic and demographic factors, then it is not surprising that they have lower uses of ICTs. Their lower use can be attributed to cumulative social disadvantage. Although there is little doubt that race has influenced access to and use of ICTs, little attention has been paid to more fully exploring how disadvantage can be mitigated by engaging in activities in the online environment, which we can describe with the digital practice model. Groups of people decide on how they will use the internet based on what is useful in their everyday lives.  As such, these practices need to be examined more fully to better understand how different groups of people use ICTs.

    People join sites because their interests are met and objectives accomplished through their use of the sites.  These networks help them solve specific tasks and individual benefits are gained. Individuals are embedded within the structural constraints of their society, and their shared relationships to the external environment shape their interactions online. Common identification among people and their commonality of experiences can attract members to remain on an online site. They share affective activities (shared beliefs, values and experiences). Interpersonal relations develop (friendships, mutual understandings) as well as functional, instrumental relations (sharing information about employment opportunities).



Appendix A:  Survey Questions and Summary Statistics for Independent Variables

General Activity
 
Have you ever created your own profile online that others can see, like on a social networking site like MySpace, Facebook or LinkedIn.com? 
(Response:  29% yes; 71% no, N=1,553)

You said you have a profile online at a social networking web site such as MySpace, Facebook or LinkedIn.com.  How many social networking web sites do you currently have a profile on? (Responses were combined into "One" – 56% and "Multiple" – 44%).  Result: 54% one; 29% two, 8% three, 5% for our more, 2% don't know, 2% refused, N=328
   

---

Affective and Instrumental Activities

What are the different ways you use social networking sites?  Do you ever use those sites to…

Based on those who have a profile on a social networking web site [N=328]
 
                                                                                                yes       no      
a.    Make new friends                                                        49        50      
b.    Stay in touch with friends                                          89       10      
c.    Make plans with your friends                                    57       43      
d.    Make new business or professional contacts      28       72      
e.    Promote yourself or your work                                 28       71     

---

References

Baym, N. 2010.  Personal Connections in the Digital Age. Malden, MA: Polity Press.

Byrne, D. 2008. "Public Discourse, Community Concerns, and Civic Engagement: Exploring Black Social Networking Traditions on BlackPlanet.com." Journal of Computer Mediated Communication 13(1): 319-340.

Campbell-Grossman, C., Hudson, D., Rebecca Keating-Lefler, R., & Heusinkvelt, S. 2009. "New Mothers Network: The Provision of Social Support to Single, Low-income, African American Mothers Via E-mail Messages." Journal of Family Nursing 15: 220-236.

Haddon, L. 2006. "The Contribution of Domestication Research to In-home Computing and Media Consumption." The Information Society 22(4): 195- 203. 

Hargittai, E. & Walejko, G. 2008. "The Participation Divide: Content Creation and Sharing in the Digital Age." Information, Communication and Society 11(2): 239-256.

Kraut, R., Brynin M., & Keisler, S. 2006. Computers, Phones, and the Internet: Domesticating Information Technology. New York, NY: Oxford University Press.  

Mehra, B., Merkel, C. & Peterson Bishop, A. 2004. "The Internet for Empowerment of Minority and Marginalized Users." New Media & Society 6: 781-802.  

Pariser, E. 2011. The Filter Bubble: What the Internet is Hiding From You. New York: Penguin.

Pew Internet and American Life. 2008. "Spring Tracking Survey." (http://www.pewinternet.org/Reports/2008/Podcast-Downloading-2008/Questions-and-Data/Spring-Tracking-Survey-2008.aspx) Accessed January 27, 2011.

Song, F. 2009. Virtual Communities: Bowling Alone, Online Together. New York, NY: Peter Lang Publishing. 

Stern, M. & Adams, A. 2010. "Do Rural Residents Really Use the Internet to Build Social Capital? An Empirical Investigation." American Behavioral Scientist 53(9): 1389-1422.
 
van Dijk, J. 2005. The Deepening Divide: Inequality in the Information Society. London, UK: Sage Press.

Warschauer, M. 2004. Technology and Social Inclusion: Rethinking the Digital Divide. Cambridge, MA: MIT Press. 

Zittrain, J. 2008. The Future of the Internet and How to Stop It. New Haven, CT: Yale University Press.