The Pheu Thai government's policy of giving "one tablet PC to every student" in school has been widely discussed in terms of both the financial cost to the country and its impact on the young recipients. The prevailing view is not to rush into the impleme
New research carried out by my colleague Dr. Jirawat Pampiemas and me in a quasi-experimental evaluation shows a potentially adverse impact on the academic performance of Thai students using computers, as measured by the OECD Programme for International Student Assessment (PISA) of mathematics and science scores.
Our sample consists of a nationally representative sample of 6,192 fifteen year-old Thai students from the PISA 2006 survey.
We specified three well-defined types of computer usage or treatment as follows:
(i) frequent usage for only educational purposes (such as using educational software, spreadsheet, or browsing the Internet for information about people, things, or ideas - 945, (ii) frequent usage only for entertainment purposes (as in playing games, or downloading games and music - 351), and (iii) frequent usage for both (i) and (ii) above 1,473. Our control group (iv) comprises the remaining 3,423 students who reported occasional or zero use for these purposes.
Having clearly defined the treatment and control conditions, we proceeded to estimate the average impact on test scores of the three treatments relative to the control outcome.
As is standard for this type of research design, the test score impact estimates were adjusted for between-group differences in the personal factors of the children, such as their family income, their parents' educational attainments and occupations, as well as the type of school they attended, their class sizes, student-teacher ratio, and teacher qualifications.
In effect, the students in the four groups described above were statistically adjusted to make them compatible with the stated observable dimensions for meaningful comparisons to be made.
Our results show strong evidence that differences in the impacts of computer use result from the type of usage.
We find that playing computer games frequently results in an average negative effect of about 16 points for science and 11 points for mathematics.
To put it another way; for a median student these are equivalent to around a 10 percentile drop in science and a 6 percentile drop in mathematics in the national ability distributions. In contrast, we find that using computers for educational purposes has small, but positive effects on student performance (between one to two points for both subjects). The impact estimates are also not statistically significant.
Our evaluation results for Thailand are very much in line with international evidence showing the impact of computer use on student achievement. These findings should raise serious concerns over the appropriateness of the government's "one tablet per child" policy, especially when our data suggests that a large proportion of students regularly use computers for entertainment purposes.
Clearly, the 'one tablet PC per child' policy should not be rushed through if we are to protect the integrity of our education system. Computers can be an invaluable aid to education if used constructively.
The least the government should do is to embark on evaluation studies on the effectiveness of the tablets and the accompanying educational software and, if necessary, begin with a pilot project. Hasty mass distribution of tablet PCs could cause untold harm to our Thai education system which needs improving in any number of ways.
1 The government has prepared a Bt3-billion budget for the 2012 fiscal year for the initial phase of the policy. Implementation will begin with the handing out of tablet PCs to every student attending Grade 1.
DILAKA LATHAPIPAT, PHD
Thailand Development Research Institute