Tuesday, May 13, 2014

Why edtech works: a primer.

Last year, one of my editors asked me whether there was any evidence that edtech works. I ended up doing quite a bit of research. But my summary email was ultimately boiled down to a single paragraph in a piece. All the rest remained in my archives until I stumbled across it today.

I wondered to myself why is this useful overview gathering dust and decided to put it to good use. Besides the evidence outlined here, one should remember that there are some good supporting theories for why edtech should be useful in the classroom and for child learning more generally. One is that it allows for ability streaming. Another is that children and teachers can get instant feedback on what they do not understand and can course-correct straight away. Another is in remediation: online videos allow material to be reviewed many times, at a child's own pace, giving many opportunities to catch up when one has fallen behind. Ultimately edtech is a force multiplier in the classroom. Teachers know this.

Anyone interested in this area should make sure to check DOE's What Works Clearinghouse and also head to the independent body that has been charged with making sure that this sort of thing works in schools. It is called Digital Promise and its Proof Points website appeared after I had composed this summary blurb. Digital Promise also produced the graphic above showing the challenges faced by its League of Innovative Schools.

Five paragraphs on edtech research

Research shows that in the classroom edtech can be an effective tool when integrated thoughtfully into the classroom. West Virginia conducted a long-term statewide learning technology program during 1996 and collected data from all fifth-grade children and found it had a positive impact on student achievement and were responsible for 11% of the gains in math, reading and language arts skills. Missouri has a program to integrate technology into instruction called eMINTs. Many years of research has shown that children in this program, of varying ages, significantly outperform those that are not in mathematics, science and social studies. And as early as 1998 a study found that higher test scores of fourth grade students in the National Assessment of Educational Progress when using computers for maths learning games. However, students who spent more time on computers scored slightly lower--which means that the way the technology is used is far more important than how often students are using it. This was underscored in 2005 when another study showed that fourth grade students using computers to edit papers did better on the English and language state tests, worse if they used these computers to create presentations. 

The Department of Education has looked at the issue too. In 2009, Barbara Means conducted a meta-analysis by screening through more than a thousand empirical studies of online learning finding 51 studies that could be tested. On average online learners performed better than those receiving face-to-face instruction but combining online and face-to-face instruction was best of all. Critics point out that none of the studies involve K-12 students, however this work has found that technology was effective across a wide age range of learners and the reason for its efficacy was to give learners control of their interactions and prompting learner reflection. Other meta-analyses from John Hopkins have found small positive effects for maths education technology across 57,000 students; positive but modest for struggling readers in 7,000 students and small positive effects for 60,000 mainstream readers. However, the most innovative technology applications showed promising evidence. 

In recent years, though, the technology has been improving at a dramatic rate, leading to much hand-wringing about the sufficiency of the research. Nonetheless, some bits of software, particularly when combined with a program of instruction, have been getting great results as determined by the DOE's What Works Clearinghouse. READ180, for example, is effective at combating adolescent illiteracy; a year on the I CAN LEARN computer based curriculum, generated gains equivalent to moving from 50th to 57th percentile; the Cognitive Tutor has had great success with its web-based artificial intelligence software that identifies weaknesses and strengths in maths. A randomised controlled trial in algebra showed that students using this tutor could reach the same level of performance in 12% less time. This is because the software adapts to the child and can assess when they are ready to receive the next piece of learning. 

This sort of learning is causing particular excitement because the literature suggests that differentiated learning in children (aka streaming) is critical to success in the classroom in some subjects even if it has been seen as not politically correct for many years. The range of abilities within an individual classroom is so great, that it is only by breaking students into groups of differing ability that teachers can manage a class. One-on-one computing with personalised learning software or "playlists" where a teacher has assembled a set of individual educational tasks for each child on their tablet, would extend this idea further. Emerging data from the roll-out of iPads is showing that children in Kindergarten show an in increase in their performance on their level of phonemic awareness and ability to represent sounds with letters. In another study, researchers gave 90 children mobile devices loaded with educational literacy apps, after two weeks one app in particular had increased vocabulary scores more than 20%. 

What has given education technology legs in recent years has been the swift success of some pioneering schools which have plunged into technology use in a big way and catalysed a shift in attitudes. Rocketship education opened a blended-learning elementary school in San Jose in 2007, where most of the students were English language learners and eligible for free or reduced-price lunches, the school was the top ranked elementary school for low-income students in San Jose and Santa Clara County. (It outperformed the more upper-crust Palo Alto Unified School District.) Children spent time in learning labs with adaptive mathematics and reading programs and the school was also able to cut its teaching staff. 

Another example is in Arizona. Charter school Carpe Diem moved technology into the classroom and its results soared while it reduced costs. In 2010 ranked first in its county in student performance in mathematics and reading. But it is the traditional public school in Mooresville, North Carolina, that is the symbolic success story. It has little money, ranking 100th out of 115 districts in North Carolina in terms of the dollars it has per student. So it gave 4,400 children in the district laptops three years ago. In three years, its graduation rate went up from 80% to 91%, and its proficiency in mathematics, science and reading went up from 73% to 88%. It is now third in the state in terms of test scores and second in graduation rate. Even the National Association for the Education of Young Children has endorsed technology and interactive media as tools in learning for young children. 

None of this should really be surprising. Other findings are clear that technology can encourage a wide range of useful things such as co-operative learning, increased engagement, decrease absenteeism, improved self esteem and motivation. Critics argue that there is not enough research on these new tools but there is way more evidence than there is for traditional printed curriculum materials. And it is fairly intuitive that technology has the potential to add to the classroom as it can do things that the teacher cannot: it is always available, has infinite patience, and can be used to identify and fill gaps in knowledge in an entire classroom. 


1. http://www.crpe.org/sites/default/files/pub_ch3_hfr12_may13.pdf
2. gains and computers at home, http://www.sole-jole.org/12192.pdf
3. phonemic awareness and ipads, https://s3.amazonaws.com/hackedu/Adv2014_ResearchSum120216.pdf
4. 11% of the gains in math, reading and, http://www.techknowlogia.org/TKL_Articles/PDF/463.pdf
5. emints http://www.emints.org/wp-content/uploads/2011/07/summary_emints_research.pdf
6. means meta-analysis line and face-http://ifap.ru/library/book440.pdf, Evaluation of Evidence-Based Practices in
Online Learning: A Meta-Analysis and Review of Online Learning Studies (also a version of this report updated in 2010)
7. Johns Hopkins website.
8. Ipads - link from amplify
9. named software, all from What works clearinghouse
10. Rocketship and the other school PDF I have
11.  http://www.nytimes.com/2012/02/13/education/mooresville-school-district-a-laptop-success-story.html?pagewanted=all