Wednesday, March 13, 2013

Introduction


Hello, welcome to my blog! In this blog I will explore the relationships between ethnicity, education and internet usage.  In this blog you will be able to see relevant studies and research on my topic, how I collected and analyzed my data, and my conclusions on the topic.

I chose my topic because most people assume that internet usage as a positive correlation with education, and I was curious as to see how race might play a factor into internet usage.  The questions I have asked are: “Do people with higher educations use the internet more than people with less education?” and “Do people of different races use the internet differently?”  I think that it is important to understand how different groups of people use the internet, especially since the internet dominates our culture.

Below please find the following:
  • Introduction to Project
  • Annotated Bibliography
  • Data, Methods, and Ethics
  • Analysis
  • Conclusions, Reflections, and Data Download

Annotated Bibliography

Cotten, S. R., and S. M. Jelenewicz. "A Disappearing Digital Divide Among College Students?: Peeling Away the Layers of the Digital Divide." Social Science Computer Review 24.4 (2006): 497-506. SAGE Journals. Web. <http://ssc.sagepub.com/content/24/4/497.abstract>. 
In this article, the researchers look into the digital divide in a university setting.  The question behind the study was:  does a digital divide exist at universities, since all university students all have access to the internet.  It turns out that it does not, but that there are differences in the ways that people of different races use the internet and differences in ways people with different amounts of digital experience use the internet.

Drori, Gili S., and Yonk S. Jang. "The Global Digital Divide: A Sociological Assessment of Trends and Causes." Social Science Computer Review 21.2 (2003): 144-61. SAGE Journals. Web. <http://ssc.sagepub.com/content/21/2/144.abstract>.

In this article, the authors chart the global digital divide between countries and blocks of countries.  They also assess the characteristics that contribute to the level of connectedness of different nations.  They compare certain social conditions in countries to the level of connectedness in these countries.  They found that it is the cultural features that cause the differences, not political or economical characteristics.

Fuchs, Christian. "The Role of Income Inequality in a Multivariate Cross-National Analysis of the Digital Divide." Social Science Computer Review 27.1 (2008): 41-58. SAGE Journals. Web. <http://ssc.sagepub.com/content/27/1/41.abstract>.

In this article, the author researches the relationship between income inequality and digital inequality and compares this to other factors that influence the digital divide.  The author used 11 variables and obtained data from 126 countries.  All the data is from the year 2005.  The author found that income inequality is a large influence on the digital divide.  He also found that the digital divide is caused by extremely complex factors.

Hoffman, Donna L., and Thomas P. Novak. "Bridging the Racial Divide on the Internet." Science 280.5362 (1998): 390-91. Science. Science. Web. <http://www.sciencemag.org/content/280/5362/390>.

In this article the authors present data about the differences in internet use between white Americans and African Americans in the United States.  They found that white people are significantly more likely to have a computer at home and slightly more likely to have one at work.  They also discuss possible causes of the disparity.

Horrigan, John. "Pew Research Center's Internet & American Life Project." Wireless Internet Use. Pew Research Center, 22 July 2009. Web. 13 Mar. 2013. <http://www.pewinternet.org/Reports/2009/12-Wireless-Internet-Use.aspx>.

The author discuss the results of a survey conducted by the Pew Research Center’s Internet and American Life Project.  The survey found that 56% of American adults have access to wireless internet, and that the most frequent way to connect was via a laptop (39%).  It also found that the mobile internet use is on the rise.

Jones, Sydney, and Susannah Fox. "Pew Research Center's Internet & American Life Project." Generations Online in 2009. Pew Research Center, 28 Jan. 2009. Web. 13 Mar. 2013. <http://www.pewinternet.org/Reports/2009/Generations-Online-in-2009.aspx>.

In this report the author discuss the different generations (age groups) that are online.  It found that Generation X is most likely to bank, shop, and search for health information on the internet.  The report also goes into detail about email use, social network use, online shopping, online banking, video downloads, and travel information with regard to who uses them.  It also shows how broadband access has increased from 2005 to 2008 by age group.

Korgen, Kathleen, Patricia Odell, and Phyllis Schumacher. "Internet Use Among College Students: Are There Differences By Race/ethnicity?" Electronic Journal of Sociology (2001): n. pag. The Socjournal. Electronic Journal of Sociology, 2001. Web. 13 Mar. 2013. <http://www.sociology.org/content/vol005.003/korgen.html>. 

In this report the authors discuss how internet use varies in race, age, and time spent studying among college age students who all have access to the internet.   The report also includes analysis of students who have computers at home and those who don’t.  The study showed that the digital divide does exist in colleges and universities.

Lenhart, Amanda. "Pew Research Center's Internet & American Life Project." Who's Not Online. Pew Research Center, 21 Sept. 2000. Web. 13 Mar. 2013. <http://www.pewinternet.org/Reports/2000/Whos-Not-Online.aspx>.

In this report the author discuss those who are not online.  The study found that while there is a digital divide caused by income differences, there is also a large gap between young people who use the internet and older people who do not.  The report also discusses those who say they don’t want to get internet access, concerns of non-connected people about the internet, gender, race and ethnicity of those who are not connected, the rural lack of access, and those who were online and now are not.  Unfortunately only a few races are mentioned here, but it was enough.  This is the report that I used to compare to my data.  I was (and still am) extremely interested in equality on the internet. 

Min, Seong-Jae. "From the Digital Divide to the Democratic Divide: Internet Skills, Political Interest, and the Second-Level Digital Divide in Political Internet Use." Journal of Information Technology & Politics 7.1 (n.d.): 22-35. EbscoHost. Web. <http://0-search.ebscohost.com.sculib.scu.edu/login.aspx?direct=true&db=a9h&AN=47906552&site=ehost-live>. 

The author of this article discusses second level digital divides, divides that vary in internet use as opposed to internet access.  The author found that second level divides are caused by motivation and internet skills.  In addition the author discusses the differences between people who are actively political on the internet and those who are not.  The author found that those who were engaged in politics online tended to have more internet skills.

Rainie, Lee, Mary Madden, Angie Boyce, Amanda Lenhart, John Horrigan, Katherine Allen, and Erin O'Grady. "Pew Research Center's Internet & American Life Project." The Ever-Shifting Internet Population: A New Look at Internet Access and the Digital Divide. Pew Research Center, 16 Apr. 2003. Web. 13 Mar. 2013. <http://www.pewinternet.org/Reports/2003/The-EverShifting-Internet-Population-A-new-look-at-Internet-access-and-the-digital-divide.aspx>.

The authors of this report (which is from 2002, so it is kind of old) discuss the “fluidity of internet use,” which includes intermittent users, drop outs, and people who have access to the internet but do not use it.  The report shows that fear of lack of skill, cost, and physical access are the most influential factors in not using the internet.  The report is divided into seven sections: who’s not online, why people don’t go online, what offline people think about the internet, the intentions of offline people, discussion of the previous groups of offline people I mentioned earlier, other social factors related to being offline, internet access with regard to disabled people (physical access is a major problem for them).  The most notable finding was the fluidity of the internet I mentioned earlier.

Data, Methods, and Ethics


Data
Our class create a common questionnaire.  We considered many possible uses and views of internet use.  We wanted to find out many things about why, how and where people use the internet and how they feel about their usage and other’s usage of the internet.  We gathered information on demographic characteristics: gender, race/ethnicity, marital status, age, education, work status, and where people lived.  We also asked questions about people’s internet use and thoughts.  These questions included but were not limited to: devices owned, personal usage frequency, personal usage place, and their expectations of the usage of the internet by others.  Finally we asked open-ended (qualitative) questions related to the internet.  These included but were not limited to: feelings of normality, social network usage, pressures to go online, and trust involving sensitive information and the internet.

By building the common questionnaire, I learned that every answer had to contain the complete list of all possible answers.  In addition these answers had to worded in a sensitive way that respected all people’s beliefs and feelings.  At the same time we had to work hard to keep the questionnaire short because respondents would fatigue with longer questions.  By keeping the questionnaire as short as possible we hoped respondents would answer all questions and not drop out.  To make sure the questions were clear, we considered many ways to word the questions.  After considering different wording we used the best possible formulation for clarity.  In each case, we tried to avoid bias by using neutral language.

We each needed to find 2 people under 30 (minimum age of 18), 2 people over 30, and one of any age (over 18).  We asked each respondent the same questions and entered the data on the online questionnaire on SurveyMonkey.

Methods

Because I am interested in equality in both education and ethnicity, I choose to analyze three questions.  All three of my questions are quantitative.  The questions are:

Question 1: What race or ethnicity do you consider yourself? (Select all that apply)
a) Black or African American
b) Native American
c) Asian
d) Latino or Hispanic
e) Native Hawaiian and Other Pacific Islander
f) Other

Question 2: What is the highest degree or certification that you hold?
a) Grade school, Jr High, or Some High School
b) High School or GED
c) Currently College Student
d) Some college (no degree), not currently enrolled
e) AA
f) Vocational Certificate or Training
g) BA or BS
h) Graduate degree: MA, MBA, JD, PhD, etc.

Question 3: How often do you use the internet on a computer/laptop/iPad.  At home? At work? At School? (Three subquestions)
a) Multiple times per day
b) Once per day
c) Multiple times per week
d) Once per week
e) Multiple times per month
f) Once per month
g) Less often

I chose these questions because previous studies (Lenhart, 2000) indicated that white people were statistically more likely to have access to the internet than black and hispanic people and that there was a positive relationship in education level and access to the internet.  I hypothesis that my data will show an overall increase in usage, as well as a greater increase in minority internet usage.

I then coded my data and removed demographics with small sample sizes.  It would not make sense to extrapolate information from such a small number of individuals.

Question 1
I removed any category with a population of less than five.  I feel that extrapolating data from such a small population would be inaccurate.
I removed any non-responses.
I combined subjects who responded with more than one race.  The two are “White or European American” and “Asian” (White/Asian) and “White or European American” and “Latino or Hispanic” (White/Latino).

Question 2
I removed any category with a population less than three.

“Some college (no degree), not currently enrolled” was shortened to “Some college”


Question 3
I have replaced blank entries with “n/a”.  I am assuming that people who did not respond to these subquestions do not currently work or go to school.  In the graphs I removed the “n/a” responses to allow for an easier comparison.

One subject was removed for not responding to any question.

Using this information I generated percentages to analyze the data.

Ethics

In order to be sensitive, we carefully considered how the questions would make people feel.  Every respondent was read the following informed consent text:

“I am requesting your participation in an interview of approximately thirty minutes for a class project. The questions will include background information such as age and gender as well as questions about political beliefs, news and media consumption, and values.
The results of the research study will be part of my final research project for a class I am taking this term. The information will only be used in this class by other students and will be published on a website for my final project. At the end of the study, the data will be erased and the website will be removed.
Your participation in this study is voluntary. All information is anonymous. No names or identifying information is being collected or will be used. Pseudonyms will be used for all data collected. If you do not wish to answer some or all of the questions you are free to say "pass" or "next question" or "decline to state" at any time. At any time you may stop participating, change your mind about answering any questions, and/or withdraw from the study altogether. You may stop the interview at any time, for any reason.
Do you consent to participate? (YES answer necessary to proceed). Thank you!”

We considered the following ethical concerns: making sure that respondents knew they did not have to answer every question, obtaining informed consent, and removing all identifying information.  We respected anonymity.

Analysis


Please note that all percentages in text are approximated.  Refer to charts for more accurate percentages.
Also note that "Less Often" refers to less often than once per month.

I must ignore AAs and Vocationals because of their tiny sample size.  I have left them in the graphs out of curiosity.

After researching and reviewing my data, I created these six graphs to represent the results of what I have found.

Internet use at Home

Figure 1: Graphical representation of internet use at home by education

Education at Home
Full Size


This graph gives a representational view of the break down regarding the use of the internet based on level of education.  There are eight sections: n/a, Never, Less often, Once per month, Multiple times per month, Once per week, Multiple times per week, Once per day, and Multiple times per day.
Associates degrees (AA) were 33% likely to use the internet at home multiple times per week, and 67% likely to use the internet once per day at home.  People with a BA or BS are 14% likely to use the internet once per day at home, and 86% likely to use it multiple times per day at home.  Current college students are 1.3% likely to use the internet multiple times per month at home, 2.6% likely to use the internet once per day at home and 96% likely to use the internet at home multiple times per day.  People with graduate degrees are 5.3% likely to use the internet at home once per day and 94.7% likely to use the internet at home multiple times per day.  People with a high school education or GED are 1.3% likely to use the internet “less often,” .67% likely to use the internet at home once per month, 13% likely to use the internet at home multiple times per week, and 67% likely to use the internet at home multiple times per day.
People with some college education are 20% likely to use the internet at home once per week, 10% likely to use the internet at home once per day, and 70% likely to use the internet at home multiple times per day.  People with a vocational certificate or training are 33% likely to use the internet at home multiple times per week, and 67% likely to use the internet at home once per day.

From this graph of multiple times per day use I have found that college students use the internet the most frequently, followed by people with a graduate degree and people a BA or a BS.  It is almost a given that college students will use the internet the most since most grew up as “digital natives.”  Trailing behind are people with some college education and people with a high school education or GED.  This follows the trend of more education, more internet use.

Figure 2: Graphical representation of internet use at work by education

Education at Work
Full Size


This graph gives a representational view of the break down regarding the use of the internet based on level of education.  There are eight sections: n/a, Never, Less often, Once per month, Multiple times per month, Once per week, Multiple times per week, Once per day, and Multiple times per day.
Associates degrees (AA) were 100% likely to use the internet at work multiple times per day.  People with a BA or BS were 3.8% likely to never use the internet at work, 3.8% likely to use the internet at work “less often,” 7.7% likely to use the internet at work once per day, and 84.6% likely to use the internet at work multiple times per day.  Current college students were 22% likely to never use the internet at work, 4.4% likely to use the internet at work “less often,” 8.8% likely to use the internet at work once per week, 17.6% likely to use the internet at work multiple times per week, 8.8% likely to use the internet at work once per day, and 38% likely to use the internet at work multiple times per day.  People with graduate degrees were 6.7% likely to never use the internet at work, 6.6% likely to use the internet at work multiple times per week, 6.6% likely to use the internet at work once per day, and 80% likely to use the internet at work multiple times per day.  People with a high school education or GED were 58.3% likely to never use the internet at work, 8.3% likely to use the internet at work “less often,” 8.3% likely to use the internet at work multiple times per week, and 25% likely to use the internet at work multiple times per day.  People with some college education were 11% likely to never use the internet at work, 11% likely to use the internet “less often,” 11% likely to use the internet at work multiple times per week, and 66.7% likely to use the internet at work multiple times per day.  People with a vocational certificate or training were 33.3% likely to never use the internet at work and 66.7% likely to use the internet at work multiple times per day.

Of the people who use the internet multiple times per day, BA or BS people used the internet at work the most.  They were followed by graduate degrees, which was followed by people with some college, which was followed by current college students, which was followed by high school or GED people.  The difference between BA or BS people and people with graduate degrees was less than 5%, which is pretty small.  College students were lower than I had anticipated, but there may be some error in the results.  People may have checked “Never” instead of not checking anything since they don’t work.  However the general trend of education being positively correlated with more frequent internet use still stands.

Figure 3: Graphical representation of internet use at school by education

Education at School
Full Size


This graph gives a representational view of the break down regarding the use of the internet based on level of education.  There are eight sections: n/a, Never, Less often, Once per month, Multiple times per month, Once per week, Multiple times per week, Once per day, and Multiple times per day.
Associates degree (AA) were 100% likely to never use the internet at school.  People with a BA or BS were 40% likely to never use the internet at school, 5% likely to use the internet at school once per day, and 55% likely to use the internet at school multiple times per day.  Current college students were 3.9% likely to never use the internet at school, 2.6% likely to use the internet at school “less often,” 2.6% likely to use the internet at school multiple times per month, 13% likely to use the internet at school once per week, 3.9% likely to use the internet at school multiple times per week, 5% likely to use the internet at school once per day and 80% likely to use the internet at school  multiple times per day.  People with graduate degrees were 85.7% likely to never use the internet at school and 14.3% likely to use the internet at school multiple times per week.  People with some college education were 50% likely to never use the internet at school and 50% likely to use the internet at school multiple times per day.  People with a vocational certificate or training were 100% likely to never use the internet at school.

This graph is potentially riddled with errors.  I fear that people checked “Never” because the internet was not around when they were in school.  I do not think this data is accurate.

Figure 4: Graphical representation of the internet use at home by race



Race at Home
Full Size


This graph gives a representational view of the break down regarding the use of internet based on race.  There are eight sections: n/a, Never, Less often, Once per month, Multiple times per month, Once per week, Multiple times per week, Once per day, and Multiple times per day.
Asian people were 4.76% likely to use the internet once per day at home and 95.24% likely to use the internet multiple times per day at home.  Black people were 100% likely to use the internet at home multiple times per day.  Latino or hispanic people were 4.34% likely to use the internet at home multiple times per month, 8.70% likely to use the internet at home multiple times per week, 26% likely to use the internet at home once per day, and 60.9% likely to use the internet at home multiple times per day.  White and Asian people were 100% likely to use the internet at home multiple times per day.  White and Latino people were 12.5% likely to use the internet at home “less often,” 1.25% likely to use the internet at home once per day, and 75% likely to use the internet at home multiple times per day.  White or European American people were 1.1% likely to never use the internet at home, 1.1% likely to use the internet at home “less often,” 2.2% likely to use the internet at home once per week, 2.2% likely to use the internet at home multiple times per week, 4.4% likely to use the internet at home once per day, and 89.01% likely to use the internet at home multiple times per day.

From the graph of multiple use per day I have found that White/Asian people and Black people are the most likely (100%) to use the internet at home multiple times per day.  Following that are Asian people, White or European Americans, White and Latino people, and Latino or Hispanic people.

Figure 5: Graphical representation of the internet use at work by race

Race at Work
Full Size


This graph gives a representational view of the break down regarding the use of internet based on race.  There are eight sections: n/a, Never, Less often, Once per month, Multiple times per month, Once per week, Multiple times per week, Once per day, and Multiple times per day.
Asian people were 28.6% likely to never use the internet at work, 4.8% likely to use the internet at work “less often,” 9.5% likely to use the internet at work multiple times per week, 9.5% likely to use the internet at work once per day, and 47.6% likely to use the internet at work multiple times per day.  Black people were 40% likely to never use the internet at work, 20% likely to use the internet at work once per week, and 40% likely to use the internet at work multiple times per week.  Latino or Hispanic people were 22.7% likely to never use the internet at work, 9% likely to use the internet at work once per week, 4.5% likely to use the internet at work multiple times per week, 9% likely to use the internet at work once per day, and 54.5% likely to use the internet at work multiple times per day.  White and Asian people were 100% likely to use the internet work multiple times per day.  White and Latino people were 12.5% likely to never use the internet at work, 12.5% likely to use the internet at work multiple times per week, 25% likely to use the internet at work once per day, and 50% likely to use the internet at work multiple times per day.  White or European Americans were 16% likely to never use the internet at work, 4% likely to use the internet at work once per week, 12% likely to use the internet at work multiple times per week, 4% likely to use the internet at work once per day, and 57.3% likely to use the internet at work multiple times per day.

From the graph of multiple user per day I have found that White/Asian people use the internet at work the most.  Following them are White or European American, Latino or Hispanic people, White/Latino people, and Asian people.  Black people were at a 0% in this category.

Figure 6: Graphical representation of the internet use at school by race

Race at School
Full Size


This graph gives a representational view of the break down regarding the use of internet based on race.  There are eight sections: n/a, Never, Less often, Once per month, Multiple times per month, Once per week, Multiple times per week, Once per day, and Multiple times per day.
Asian people were 5% likely to never use the internet at school, 5% likely to use the internet at school “less often,” 5% likely to use the internet at school multiple times per month, 5% likely to use the internet at school once per day, and 78% likely to use the internet at school multiple times per day.  Black people were 20% likely to use the internet at school once per week and 80% likely to use the internet at school multiple times per day.  Latino or Hispanic people were 42% likely to never use the internet at school, 5% likely to use the internet at school once per day, and 53% likely to use the internet at school multiple times per day.  White and Asian people were 33% likely to use the internet at school multiple times per week and 67% likely to use the internet at school multiple times per day.  White and Latino people were 16.7% likely to never use the internet at school, 16.7% likely to use the internet at school multiple times per month, 16.7% likely to use the internet at school once per day, and 50% likely to use the internet at school multiple times per day.  White or European Americans were 24% likely to never use the internet at school, 1.35% likely to use the internet “less often,” 4% likely to use the internet at school multiple times per week, 4% likely to use the internet at school once per day, and 66% likely to use the internet at school multiple times per day.

From the graph of multiple use per day I found that Black people are the most likely to use the internet at school, closely followed by Asian people.  Following them are White/Asian people, who are closely followed by White or European Americans.  Following them are Latino or Hispanic people, and White/Latino people.

Conclusion


After spending time reviewing and analyzing the data I have come to the conclusion that education does correlated with frequent (multiple times per day) home internet use (disregarding college students who are most likely digital natives).  This shows that there is still a level of digital divide in our society, and that it is education based.  It should be noted that in 2000 only 34% of High School grads, 63% of people with some college, and 75% of people with a college degree or higher had access to the internet.  This has dramatically increased over the past 13 years according to my data, at least in home internet access.  Everyone in my refined data set had access to the internet at home, and everyone in the original data had access to the internet at home except for one person (who could have just left it blank by mistake).  I have also come to the conclusion that there is a similar correlation of internet use at work based on education level, although there is the possibility of data entry error here.  I am concerned that people would often check “Never” when they meant to not check anything at all.  This fear is even more prevalent in internet use at school as many of the older respondents went to school when the internet did not exist (or at least in a form similar to what it is now).  I could have removed the “Never” section and made a graph of frequency of internet use of people who use the internet.  However, if I had done that I would have run the risk of having a small sample size.

My other conclusion is that there is still a gap between race and internet use at home.  Looking at the study from 2000 (Lenhart, 2000), everyone (white, black, and hispanic/latino) has become more connected in the past 13 years.  This gap is not in overall use, but in frequency of use.  From my results I found that black people were 100% likely to use the internet multiple times per day at home, but they had a sample size of 5 so I would not (and you should not) conclusions from such a small amount of data.  The mixed race sample sizes were also under 10, so I cannot accurately draw conclusions for them either.  White, Latino/Hispanic, and Asian people had large sample sizes, so data regarding them is more accurate.  I found that Asian people are very likely to use the internet multiple times per day (95%), as are white people (89)%.  Hispanic/Latino people are only 60% likely to use the internet multiple times per day at home.  This means that connected White/European Americans have increased by 39%, where as connected Hispanic/Latino people have only increased by 16%.  There is obviously some growth gap present.  I don’t want to draw any conclusions about work and school due to the ratios of education levels and race levels for some groups.

Reflecting on this project, I have learned that the key to any research project is a very large data set.  I could not draw as many conclusions as I had hoped since the majority of data were college students and White or European Americans.  It would have been better if we had set out to survey the same number of each race/ethnicity and each education level, or at least set a minimum number of people for each category of people.  It would be even better if we had set out to find a set minimum number of people each race/ethnicity and education (IE: black college graduates, white high school/GED).  I would have been able to draw conclusions for more categories of people if we had.  I have also found that questions on surveys need to be labelled as clearly as possible because I believe that portions of my data are inaccurate.  More specifically, we should have added a selection called “N/A” or “The internet did not exist when I was at home/work/school” to Question 3 (the one about internet use).  With that one change to the survey my data would have been much more accurate.  I now know the pain of statisticians and sociologists.

If you would like to view all of the data in an Excel form, download it here.  Chart 3.0 contains my most refined data.