How to Improve Schooling

Press "Delete" and Start Again

Let's think again, please Christmas is fast approaching. You ponder what presents to delight your progeny with. Something hip, say, but also useful, wouldn't that be the ticket? Something like a computer? What with the internet and educational software and word processors and worksheet applications, your offspring couldn't but shine in school.

Well, think again.

A new study by two of Ifo's education experts, Ludger Woessmann and Thomas Fuchs, took a detailed look at the PISA data —measuring student performance in a number of OECD countries— and found that computer availability at home could actually be detrimental to educational performance, while computer availability at school was found to show no discernible positive effect.

This appears to fly in the face of conventional wisdom. After all, the initial release of the PISA findings did state —albeit dampened with a caveat regarding methodology— that a positive correlation existed between computer availability and educational performance.

What the Ifo researchers did was scratch a bit deeper under the surface by controlling for students' family background and availability of other non-computer resources at school.

Family-background characteristics have been shown by many studies to play a significant role in students' educational achievement. And presto, when you analyse the PISA data taking such background into consideration, the positive correlation between computer availability at home and performance at school turns negative. It would appear that computers at home are not exactly used for running educational software, mining the internet for useful data or composing better homework assignments, all things that would have a positive impact on performance. Rather, computers at home are used for playing games, chatting and otherwise providing entertainment. Computers thus displace other activities more conducive to learning.

At school the picture is a bit different, but the result isn't. When the availability of other resources at school is duly accounted for, the mere availability of computers does not translate into higher student performance, at least not beyond a certain frequency of use. This gives rise to a curiously inverted-U-shaped curve for performance against frequency of use: little use, poor performance; moderate use, higher performance; frequent use, poor performance. The pivotal thing seems to be what you do with a computer, not the mere availability of it.

The disappointing effect of computer availability at school appears to point to displacement of more effective teaching methods by time spent before a computer screen or, even, to diverting of funds that might have been better allocated to instructional material or improving teacher training. Reducing class size, after all, has also been found to have barely any discernible positive effect, as the same authors report in a forthcoming paper to be published by the European Economic Review.

Reforming the school system's institutional structure, on the contrary, does appear to be highly beneficial, as reported in yet another paper by Woessmann, where he shows why schools should enjoy more autonomy in the selection of teachers and textbooks, and in how they shape budget allocations within the school itself. Furthermore, the application of external exams —i.e. exams free of influence by the teaching personnel in the schools themselves— helps to ensure a better overall quality level, as reported both in the Woessmann paper and in another CESifo Working Paper authored by Hendrik Jürges, Wolfram Richter and Kerstin Schneider.

The effectiveness of the above measures is corroborated by a further finding of the research team, which shows that pupils of private schools, such as those run by churches, achieve better performance levels than do those attending public ones.

So, it is not enough for governments to pour billions into cluttering the classroom with monitors, mice and keyboards. If teachers are not properly trained, educational institutions remain unreformed and exams are not of a sufficiently high, uniform level, poorly performing countries will just keep getting a rap on the knuckles. Time and again.

T. Fuchs & L. Woessmann: Computers and Student Learning: Bivariate and Multivariate Evidence on the Availability and Use of Computers at Home and at School, CESifo Working Paper No.1321

L. Woessmann: The Effect Heterogeneity of Central Exams: Evidence from TIMSS, TIMSS-Repeat and PISA, CESifo Working Paper No.1330

Hendrik Jürges, Wolfram Richter & Kerstin Schneider: Teacher Quality and Incentives: Theoretical and Empirical Effects of Standards on Teacher Quality, CESifo Working Paper No.1296