What is ETD?

ETD promotes discussion on low-cost ICT initiatives for educational systems in developing countries. Read More

Join ETD

Become a part of the conversation. Contribute your ideas, strategies and expertise to our discussions. Join Now

Do Computers and Internet Access at Home Reduce Student Test Scores?

Wayan Vota

The recent New York Times article Computers at Home: Educational Hope vs. Teenage Reality brings into question the benefits of computer and Internet usage at home to improve education.

new york times olpc
Is this the home ICT usage reality?

The article cites two studies that indicate that when computers and Internet access are provided to students for household usage – either subsidized or not – home-based ICT does not have any measurable educational benefit, even reducing test scores in math and English in low-income students.

First, from Romanian data, we have Ofer Malamud and Cristian Pop-Eleches writing in Home Computer Use and the Development of Human Capital that:

“We collected survey data from households who participated in a unique government program in Romania which allocated vouchers for the purchase of a home computer to low-income children based on a simple ranking of family income.

Children who won a voucher had significantly lower school grades in Math, English and Romanian but significantly higher scores in a test of computer skills and in self-reported measures of computer fluency. There is also evidence that winning a voucher increased cognitive ability, as measured by Raven’s Progressive Matrices. We do not find much evidence for an effect on non-cognitive outcomes.”

Next, from North Carolina data, we have Charles Clotfelter, Helen Ladd and Jacob Vigdor writing in Scaling the Digital Divide: Home Computer Technology and Student Achievement that:

Using within-student variation in home computer access, and across-ZIP code variation in the timing of the introduction of high-speed internet service, we also demonstrate that the introduction of home computer technology is associated with modest but statistically significant and persistent negative impacts on student math and reading test scores. Further evidence suggests that providing universal access to home computers and high-speed internet access would broaden, rather than narrow, math and reading achievement gaps.

For this month’s Educational Technology Debate, we’ll question these studies, and similar ones, to try and explain these results and what they might mean to educators in the developing world.

What do you think of these reports?

So what is your opinion? Did the researchers find an ugly truth we’ve been ignoring, as this quote suggests:

“We found a negative effect on academic achievement,” Ofer Malamud said. “I was surprised, but as we presented our findings at various seminars, people in the audience said they weren’t surprised, given their own experiences with their school-age children.”

And was the implication from Jacob Vigdor accurate, that in low-income homes (vs. more prosperous ones):

Left to their own devices, adolescents may be more likely to use computers for non-productive purposes, to the potential detriment of their academic performance.

Or should we as technologists and educators reject their findings as missing the point, that we should Stop Wasting Children with ICT4E Assessments, and get on with making sure all students, from all economic backgrounds, have ICT in their schools and their homes?

Don’t be shy, share your thoughts in the comments below, or as Guest Post on this topic.


Don’t miss a moment of the action!

Subscribe now and get the latest articles from Educational Technology Debate sent directly to your inbox.

21 Responses to “Do Computers and Internet Access at Home Reduce Student Test Scores?”

  1. Studies from Uruguay and Birmingham, and reports from many other programs, suggest that XO computers are usually little used in school. (This lack of use is not a feature of other laptop programs, where the schools own the equipment and plan to use it in an integrated pedagogical/curricular approach, but is common in OLPC programs in which the computers are given to young children.) WIth little use in school, it is likely that the only substantive benefits could come from home use. I and others have questioned what evidence there is of substantive benefit to children from access to computers at home, outside of any instructional or mentoring program. Though there have been well over a million XO contributors distributed around the world, there is no study of their impact that contradicts the findings of these two other studies Wayan refers to.

  2. Any resource given indiscriminately to children will provide an opportunity cost. If you did a test on use of the XO I’d be surprised if the ones regularly using the device did not score higher than those who did not. What matters are two fundamentals 1. What do we value in terms of measuring valid learning outcomes? 2. How do we MANAGE the resources to prioritise supporting the most valued outcomes? I don’t think it takes complex studies to realise that social chat via an XO or playing a computer game is not going to improve scores in algebra or formal writing, That does not mean properly managed use of technology can not help or that learning to use technologies effectively in a technological is not important.

  3. Christpoher Segot

    I keep reading articles and reports from programs that issue technology to the students with little to no correlation back to the classroom studies and in almost every case the results are the same…. the students become more adept with computer skills but their is little to no academic improvement. I have run fairly successful 1:1 laptop programs for SES/NCLB and the key elements can be broken down to: the technology must keep the student engagedit must relate back to the classworkit must be incorporated into the curriculum, no different then assigning homework in a paper workbookthe teachers must see the benefit and ‘buy-in’if possible have some sort of home-based synchronous learning, assigning asynchronous learning assignments and then expecting all the students to be motivated to learn on their own when they could be chatting, or playing computer games is a stretch at best. Their needs to be a curriculum designed from the ground up to meet specific goals and align with specific standards that incorporates the technology in and out of the classroom and have the results monitored to determine the real effect of ICT in the learning process. Reports like these that do not have significant incorporation in the classroom give the wrong impression to the public and they only hurt the students as well as continue to waste money as long as schools buy new technology but do not properly incorporate them.

    • Mark Gutierrez

      Could you cite some of the articles of these studies?

  4. Grant Robertson

    These reports prove one thing: That simply having a computer does not educate a person any more than simply having a refrigerator will keep them from going hungry. As Christopher Segot explains better in the above comment: It’s what the students do with the computer that counts. If they just use it to play more video games or do more social networking then it hasn’t done much good. I would add to Christopher’s comments that students must be taught, convinced, conditioned, or just plain gotten interested enough to actually want to learn. William Glasser, in his 1998 book, Choice Theory, claims one of the primary reasons students don’t learn is that teachers spend more time and energy disciplining them rather than trying to get along with them. It goes against the common intuition that seems to be prevalent in schools these days but it does make sense to me. If children don’t like their teachers then they, by association, won’t like what the teacher is trying to teach. I have to disagree with Christopher in that I do not believe that simply monitoring results will improve learning. All that really does is generate more research data. Useful, yes, but not necessarily for the education of the students being monitored. Finally, learning itself must be fun. Not necessarily all video games, but the material should be presented in a way that makes it much easier for students to actually succeed at learning while still being challenged. The problem with just dumping laptops into students, well, laps is that it is about the same as locking them into a library where all the books have had all their pages torn out and piled up in the center of the room. Sure there is a lot of information out there on the internet but the student has to go to a lot of work to find it with no guarantee that they will ever find something that interests them or that suits their education level or learning style. I wrote an essay on the topic and posted it on my blog last year. If anyone is interested, you can find it at: http://demml.blogspot.com/2009/07/philosophy-stat

  5. Did the studies just show that poor children have lower achievement scores than more affluent children, or did their scores decline from previous levels? From the description in this article, you can't tell. If it's the first, well, Garbage in, Garbage Out.

    • The North Carolina report shows a drop in test scores once computers and Internet connectivity came to each zip code they analyzed. So this means that there was an actual decrease from previous years, relative to the same household income groups in prior years. The decrease was greater in less affluent homes. This has significant relevance to educational systems in the developing world where a majority of the families would be described as low-income, even relative to their own countrymen.

      • Now we are talking, Wayan! This is the ugly truth ignored so far indeed. We do not yet know how to use these new technologies to build up knowledge and skills – in the long-term memory of the pupils – systematically. On our blog we pay a lot of attention to this problem in relation to the technology as it is today. The problem is evident in rich countries who have taken the lead in digital education and already invested billions of dollars in it. Here a ‘quantum leap’ in pedagogy is needed to make it productive, see: http://visualteach.blogspot.com/p/about-this-blog… Strategical consequences on how to introduce the benefits of computer technology in educational systems in the developing world are briefly suggested in http://visualteach.blogspot.com/p/improve-educati… In conclusion: not less investments in computer technology in the educational system but a shift in priorities.

  6. robvanson

    What are we measuring in these studies? – A drop in achievement in formal (drill) skills: Math, foreign language, and native grammar/literature. – An increase in cognitive ability. Sounds like a classic example of distraction by an intellectual or cognitive challenge. The children are enticed by a new activity that engages their intellectual abilities and time so they spent less time doing their homework drills. My curiosity is raised by the boring nature of the pre-test environment of the children. Was life really that bad that the children would rather do their homework? In the long term, we can predict children with Internet access will be good readers (the Internet is still a lot of text) and generally well informed. But would their better access to native and foreign language material increase their reading, language, and other skills? Or would all gains be undone by the distraction? Now there is one big question hanging over this study. I wonder why no-one else has raised it: If children with computer+Internet at home indeed perform worse in school, then all over the world poor children should out-perform rich children in school. So why is it, that this is NEVER the case in any community? My personal guess is that the distractions of the computers are more than offset by the motivation of the parents. There is an old joke (I got it from a talk by Flynn, the version is from old New England): “Irish boy gets home with good grades, but he didn’t make the football team. Father says, good work, but the disappointment radiates through. Boy will spent more time to get on the team. Jewish boy gets home and tells he made the school top football team. Parents furious and yell: You will hurt your brain and won’t go to medical school. Boy leaves team.” These studies show the two effects of poverty on school achievement: – An environment with cognitive scarcity leading to a lag in cognitive abilities – Lack of motivation to improve school achievements leading to lower grades In such an environment, any distraction will (initially?) reduce time on homework drills and reduce grades further. The question is whether the decrease in ignorance and increase in cognitive abilities can make up for that? Or maybe we need a parallel program to get parents involved in school? (a good predictor of school success)

  7. Christopher Segot

    @Grant I would have to agree that just monitoring is not enough…. however having worked with many educators in various settings and school/district/state administrators I can tell you that they are very driven by metrics and results and they use them to gauge learning and where to allocate resources (however incorrect this is it is still a fact, to get the much needed funding we need to get them metrics) Another thought that I didn’t mention in my first post but robvanson mentioned was parent involvement. This is very crucial to the learning process but not always possible. Many of the students that I served in very populated areas of California had many issues with parent involvement, remember one of the qualifiers for NCLB is that they qualify for free or reduced lunch… meaning that the parents probably don’t earn very much and may be working 2 jobs each to keep food on the table, in addition many of the parents still do not speak English… guess what language their homework is in? (well in most districts, however standard testing is always in English) At the beginning of each program start I wold have my team hold 4 hour family technology training meetings that got the whole family involved with the technology that was going to go home with them and I found this to be very vital to the program success and while we kept the focus on aligning their laptop with the curriculum and improving academic skills it took constant involvement from the instructors to keep the students engaged and build a positive rapport with the parents. I haven’t had the opportunity to work with any of the olpc deployments in remote third world countries but from what I have heard and read from the people involved their are much bigger issues and while inherently parent involvement is vital to any academic success the programs that get implemented need to address the issue almost as if their is none at all. And of course it would be very difficult to demonstrate metrics to school administrators to tell them what they should already know but do not put a monetary value on developing programs to implement parent involvement.

  8. Learning to use computers efficiently is part of working life and most adults with good degrees don’t do it well. They are still sending each other Word and Powerpoint attachments when the same information could be indexed and linked into cloud based applications to better effect and at lower cost. There might not be a test score for IT skills but they are important in their own right – if there was such a test most adults would fail it. But teachers are important. We need to teach children how to be self-sufficient in IT using free internet based resources safely. It is not just a matter of buying them a toy to play with or lashing out on expensive software. Digital natives can so all that is needed with very little money. If you thnk education is expensive, ignorance is a lot more so and that includes IT.

  9. Saba Musharrif

    It is children that we are talking about here and as we all must know, children are always curious. For the child the computer is a new toy (and with interenet access its a whole new world of fun and games), and therefore ‘play time’ needs to be monitored. When we provide the child with a computer and internet access we are providing them with a whole bunch of options and he will chose that which interests him the most; majority of the children will find English and Mathematics the least interesting. I agree with Christpoher Segots point that the curriculum should be designed to incorporate technology in and out of the classroom. Also if we are intending to educate children through eLearning, then the applications should be so engaging that the student “enjoys” using these applications, and is not easily distracted. The internet is an excellent resource if used constructively; there is unlimited data/information to which access is gained within seconds, and it is for this reason that technology must be incorporated within the classroom and curriculum.

  10. Thomas Grotkjaer

    Don’t Jump to Conclusions Most people/academics want to draw conclusions from a new set of data. It is more difficult to publish a study where nothing can be concluded. Most journalists want/need a good story every day, and a good story is usually when something goes wrong. They have a newspaper to sell everyday. That is not to say that these three new research studies contain interesting reading. They bring to light important aspects of using scarce resources within an education sector. But let us not jump to bold conclusions right away. 1. The Romanian and North Carolina studies can be used to say… something…. about how computers were used… in Romania and in North Carolina. I believe that anthropologists would agree with me, that it is difficult to assume that other locations have the exact same context. And the local context is important for education activities. 2. Hands-up everyone: How many has actually read all the pages? They are extensive reports. I have only read the conclusion on the North Carolina study, so let me focus on the study from Romania that I have read more about. 3. How many of you would call the voucher-program in Romania an education program? Hands-up everyone! Where are the teacher involvement, school involvement, the curriculum integration, the school-parent orientations etc. like mentioned by Seget before me? 4. This voucher program is clearly (to me) a computer program but a successful computer program with 98.6% of the households apparently used the voucher to buy a computer (although it seems that the control group families more or less anyway bought a computer after they were refused a government voucher!). So the alternative cost seems low (to an economist like me) for the government, i.e. perhaps it didn’t need the voucher program? 5. Only 9% of the program households had installed the education software that is aligned with the national curriculum (I couldn’t find the corresponding figure for the control group). With little integration into daily school activities and only 15% of the households having access to the Internet… is it any wonder that many installed games on their computer and spent some time on them? 6. The study wants to measure the difference in academic achievements with two groups, the households who got the voucher and the households who didn’t get the voucher. To me these two types of households are distinctively different in terms of income. I am guessing that a story similar to the one by Robvanson’s about the Irish and Jewish boys can be found in Romania as well. At least I know that in most low-income countries there is a strong correlation between higher income and higher education of women that again correlate with higher and better education participation by their children. I haven’t read that the study is discussing this research premise very much… 7. Assuming that it is possible to compare the two groups – the study wants to find difference in academic achievements after the children have used the computers less than 12 months (approximately in the summer 2008 when they households received the voucher until May/June 2009 when the survey was done). This is not a long period for academic achievements to take root. Academic progress is a long-term process (every educationalists would tell us!), and there is no reason why this should change when using an extra(?) learning tool like a computer. What we could perhaps expect to see (if it had been an education program!!) within a period of 12 months was changes in academic behavior and perhaps attitude (i.e. that they perhaps study for longer periods of time and in a different way and that they perhaps like to study more now etc.). Let me not go on with my worries about concluding too much on the study from Romania. I would say that the study might help the Romanians to improve their support to education. I would argue that they perhaps would need to look at the objectives of the voucher program; is it a computer or education program. If it is the former, perhaps it is relatively successful program. If it is the latter, there are good reasons for the Romanian government to look critical on the program; in the current form it would be a wonder if it would achieve broad academic progress (aside from computer and ICT subjects). Aside from this, I will let NyTimes provide speculations about how computers “are used at home” by children but let’s not jump to broad conclusions that will be used across different locations and countries. Every country needs to work within it own context and conditions.

  11. I am doing some work with eight-year olds in Trinidad, West Indies. Many use their computers for school projects or for games or to search for info on areas they already have an interest in. The use of the computer for their projects are mostly done in away that in some encourage things like cutting and pasting or plagiarism.
    The computer is a tool and lessons have to be structured to teach children how to use them for research and general learning Use of it should be monitored and regulated. Many parents do not regulate the use of all these devices enough, be computers, television DS the whole lot.

  12. Here is the best place to find links to research on OLPC XOs in schools.
    http://wiki.laptop.org/go/OLPC_research The research clearly shows that dumping XOs into schools without teacher training and curriculum development matching local needs has only modest benefits and some negatives, while doing it right has not only educational but social benefits. This is in line with research over more than 50 years on computers in education, starting with IBM 360s and rooms full of terminals. The difference today is that netbook computers cost less than printed textbooks, so development of elearning materials under free licenses can improve education and will certainly save money. More at http://www.earthtreasury.org/wiki.cgi?ReplacingTe

  13. Daniel Levin

    I am sorry but the article in the NYT was very one sided – it is very clear that the answer will be part of how complex is the eco system – dose teachers training support encouraging these kids to study at home, do they challenge them to study at home, the research from Texas actually confirms this if you read between the lines.. and there is of course the parents issue…so jumping to concussion on the base of a few vouchers in Romania is a waste of time. The question is much MORE COMPLEX than this.

  14. Tony Forster

    I believe that the results of the Romanian study are invalid due to the misuse of regression discontinuity analysis. For the technique to be valid, an underlying distribution bust be linear, which it is not. I give more detail at http://tonyforster.blogspot.com/2008/06/home-pcs-

  15. Are you making money through your blog now?Well,I decide to try making money online ,but I still have many questions.Can you give me some tips?


Subscribe to ETD

Enter your email address:

Delivered by FeedBurner