TECHNOS

November 20, 2008

TECHNOS QUARTERLY Spring 2002 Vol. 11 No. 1

The Productivity of Learning Technologies: A School and Learning Policy Review

By Dale Mann

It's time to sing the productivity praises of our educational technologies.

Technology has transformed the American economy and penetrated important parts of public schooling, yet educators are curiously silent about the productivity effects of technology and schools. Both the explanation for that and the price paid are political. We slow the progress of learning and diminish support from the public because we do not discuss or celebrate gains in productivity. This article reviews a range of examples and describes the near and intermediate term prospects for productivity.

American Economy, Technology, and Productivity
Think of how the 1990s began — America was regarded as a rust-bucket economy producing too few, low-quality products for high prices. Other countries invaded and then dominated all of the continuous production processes that we had invented. Automobiles, steel, electronic assembly — all went overseas.

But the 1990s ended with America as the world's only super-power, economically as well as politically. Forty percent of the growth in the gross domestic product of the 1990s has been attributed to productivity gains enabled by technology. “Non-farm productivity rose at a healthy 2.5 percent annual rate in the quarter (Q2, 01), as businesses produced roughly the same output using less labor. … The unusual feat offers further evidence that, with the help of technology, the economy has emerged from a 20-year productivity slump …” And, “If productivity gains … reflect a country made more efficient by computers and the Internet, the long-term average growth of the economy may be about 3.5 percent” (David Leonhardt, “Productivity Still Gaining Despite Slump,” New York Times, Wed., Aug. 8, 2001, p. C1).

How does that happen? Think about the effect of technology on different sectors of the economy.

Transportation. At the beginning of the 20th Century, transportation meant trains, horses, and “motorcars.” But every car built since 1996 has more computing firepower than did the first lunar lander, Apollo 11.
Finance. Eighty percent of personal banking transactions are done by computer. Rather than wait in line for personal contact with a teller, we use an ATM. ATMs operate in a broadband network and are powered by water-cooled Crays, but learning to use one did not require any of us to go away for a weekend workshop. Banking technology is transparent; school technology is not.
Communications. When the telegraph first came to Russia, Leo Tolstoy was driving his mother down a country road. She asked what the poles and wires were for, and he replied, “They are used to move words.” After watching for a while, she announced that she had not seen any. Today, Norwegian teenagers can use their cell phones to buy a Coca-Cola from a vending machine. Eleven million young adults now use wireless; in three years, the number will quadruple to 43 million (Stephanie Miles, “Wireless Entertainment Attracts Investors,” Wall Street Journal, March 6, 2001).
Agriculture. Farmers were once legend as the most recalcitrant sector of the economy. No more. Tractors are sold with transponders that talk to global positioning satellites and adjust the amount of seed and fertilizer for every meter depending on an on-board computer analysis of the soil composition and moisture.
Business. First electronic technology and now digital technology have transformed work. Remember “typing pools”? Now we all “word-process” what was once typed by someone else. The smart organization buys its employees laptops, confident that professionalism will promptly blur the lines between company time and personal time, home and work.

American Schooling, Technology, and Productivity
Here are some examples of productivity gains in American schooling and then in education, largely conceived. Schools have made big gains. Twenty years ago, the folklore was that change in schools took 30 years. Consider the ten-year growth represented in Figures 1 and 2.

A Per Child/Per Month of Gain Example
Imagine that you are required to increase reading scores by one month in grade-level gains. There are two known methods: (1) reduce class sizes from 21 to say, 15 students per class, a technique that requires hiring more adults, or (2) add instructional technology.

According to an analysis from Lewis C. Solmon, both methods will yield the same outcome - a month's increment of learning. But the teacher method is more than seven times more expensive than the technology method. “… [T]he total cost figures … translate into $636 per student per year for class size reduction and $86 for BS/CE [computers]” (Lewis C. Solmon, “Afterword,” in Dale Mann, Charol Shakeshaft, Jonathan Becker, and Robert Kottkamp, The West Virginia Story: Achievement Gains from a Statewide Comprehensive Instructional Technology Program, Santa Monica: Milken Family Foundation, p. 49). Every school board in America claims to be interested in “cutting costs,” yet in this situation, nearly every school board would hire more adults.

A Per State Example: West Virginia
In 1991, West Virginia was fortieth among the American states by per capita income and thirty-third in student achievement. The iron link between social-class standing and school-class standing means that per capita income predicts student achievement. In other words: Children from low-income families have low achievement. In 1999, West Virginia's per capita income had not changed, but its pupil performance had moved 22 places up the list to eleventh. A study supported by the Milken Family Foundation discounted every explanation for those gains but one: the statewide press to include technology in instruction represented by the “Basic Skills/Computer Education” program (Mann et al., West Virginia).

Technology and telecommunications can stretch dollars and get to equity and ubiquity faster and with less pain.

A Teacher Professional Productivity Example
In a study of technology use in New York State, 4,041 teachers responded that technology was good for student learning but that they did not believe their school boards would buy much more of it. Interactive, Inc., who conducted the study, wondered whether teachers would be willing to accept larger classes — maximizing per classroom tax revenues — in return for more technology. They refused. So Interactive Inc. added some other incentives into the mix, and two-thirds of the teachers agreed to the following package, a deal that would leave $12,500 per classroom as a bonus for the school board:

five more students for each class (thereby increasing the then-average per-pupil expenditure of $7,500 by $37,500)
$15,000 to purchase technology for their classrooms
$10,000 salary increase

The same day the study was released to the public, the lawyer for the state's teachers called to object to the unacceptable (to him) implication that teachers were willing to pay for part of their own professional tools. But that is exactly the kind of gain-sharing that the teachers had volunteered (Dale Mann and Edward A. Shafer, “Technology and Achievement,” The American School Board Journal, 184:7, July 1997, p. 23).

A New Teacher Recruitment Example
The United States is famously assumed to need to recruit two million teachers in the next decade. The conventional method is through other face-to-face encounters and teacher fairs. New York City recruits in India (air fare $3,948), and Cleveland recruits in the Philippines (air fare $3,105). The minimum cost to have the district represented at a job fair is $1,200. The largest 100 schooling jurisdictions in this country might do 10 fairs a year, which would cost a total of $1.2 million for just those districts. But a Web-based simulation could be developed to attract, screen, orient, pre-qualify, and place prospective teachers for a mere $1.25 per candidate.

Three Hardware Examples
Thin-client computers operating in a networked environment can provide all the functionality associated with desktop computing but for about 40 percent of the existing cost. The one-to-one student-to-device ratio is a bonus. Based on an analysis of six public school districts from across the United States, Interactive, Inc. determined that the average cost to equip an elementary school classroom with five PCs and a teacher desktop is $6,869 including networking support at the classroom, building, and central office levels. Putting six Sun Ray thin-client devices in the same classroom - five for students and one for the teacher - costs $2,666 and includes that classroom's pro-rated share of a central server. (The cost of a 25-device bundle and supporting central server is $16,000.)

Wireless networking captures many of the same advantages but because wireless platforms are ordinarily hand held (iPAQs, Palms), they add the advantage of ubiquitous computing. Combining handhelds with wireless solutions offered by companies like Mindsurf Networks has the promise of 1:1 student:device equity and nomad computing, which moves learning to all its settings.

Arizona is the first state to apply an application-service provider (ASP) to all of its schools. Rather than installing individual copies of software on every desktop, 252 titles — with an additional 7,000 titles available at reduced costs — are available free to every school in the state (the state assumes the subscription costs of about $8 per student or teacher user). The “rent-not-buy” initiative fills the pipe previously created with the statewide wiring contract through QWEST.

The hardware examples demonstrate how technology and telecommunications can stretch public dollars and get to equity and ubiquity faster and with less pain.

A Dollar Recovery Example: Data Mining
One version of data mining software tracks the eligibility of students for federal Medicaid benefits. After installing 4GL Inc.'s software, Baltimore City's Medicaid recovery went from $2.5 million to $25 million; Detroit's went from $3 million to $15 million (Elizabeth Guerard, “Schools Use Software to Track Special Ed Services,” eSchool News, 4:6, June 2001, p. 1).

A Foregone Expense Example: School Dropouts vs. Phone-Card Theft
Think about a 14-year old ninth grader. She may have all the warning signs of dropping out: falling grades, dysfunctional home life, police encounters, no co-curricular involvement, too much time in paid employment. The downstream costs for not intervening will be hundreds of thousands of dollars in remedial education, increased social dependency, decreased lifetime income, and decreased taxes. Different “child serving” bureaucracies hold different data about the ninth-grader that might flag her for help, but no agency tracks her comprehensively and thus no agency intervenes.

But, if one of your credit card numbers gets used to make three phone calls to Belize, and if you have no prior history of calling Belize, your home phone will ring with a company representative asking if you authorized the calls. Falcon Inc. software (HNC Software, San Diego) and an Omaha call center start protecting the company from unrecoverable charges at $20. Falcon checks 12 million credit-card transactions a day and saves $500 million a year in fraudulent credit charges (Barnaby J. Feder, “Artificial Intelligence for the New Millennium,” New York Times, June 30, 2001, p. C1). Similar technology could be used to help students at risk.

From most to least powerful, the sources of education in a child's life are the family, the media, the peer group, the school.

An “Other Educators” Example
With state standards, accountability, and other sanctions in place, it is hard to imagine that there are many gains from pushing schools harder.

Laurence Steinberg writes in his book, Beyond the Classroom: Why School Reform Has Failed and What Parents Need to Do (Simon & Schuster, 1996): “ … [S]tudent achievement is due more to the conditions of students' lives outside of school than … within school.” He feels that we have disregarded “… forces that, while outside the boundaries of the school, are probably more influential.”

Fortunately, schools are not alone. In 1966, James S. Coleman et al (Equality of Educational Opportunity, Washington DC: U.S. Department of Health, Education, and Welfare, 1966, p. 77) documented that, from most to least powerful, the sources of education in a child's life are

·the family
the media
the peer group
the school

In 1966, “the media” meant black-and-white TV and stick figures on a computer screen. By 2005, households with broadband access will likely increase by 900 percent, from 5 million to 47 million (David Lake, “The Five-Year Forecast,” The Industry Standard, March 26, 2001, 4:12, p. 83). And children are already sending us signals about how they learn: 60 percent of the preferred sources in Figure 3 are digital.

How does this power learning? Lightspan Achieve Now! is a CD-ROM and Web-enabled K-6 reading, language arts, and mathematics curriculum, linked to state standards, launched in the classroom but (critically) completed at home. Multi-year, multi-site research indicates that for every minute that the teacher spends with Lightspan Achieve Now! in the classroom, families spend six minutes at home. Compare the cost-free service of the parent-as-educator to the cost of providing home visits by teacher aides (if you could amend their contract to require home visits).

School people can add other, practical examples to this review. Why, in the presence of evidence about the contributions of learning technology, is the topic so little discussed?

Inattention to Productivity and Teacher Unions
Schooling is the most labor intensive sector of the American economy. A higher proportion of teachers belong to labor unions than auto workers, steel workers, or coal miners. Discussing productivity is so unacceptable to teacher unions that prior to the negotiation of a New York City teacher-union contract in his first administration, even Mayor Rudy Giuliani instructed his team not to raise the issue.

Productivity measures the output per worker. What is productivity in schooling?

Increased amounts of learning by students;
Increased number of students learning …
— with the same number of adults, and/or …
— with fewer adults.

The last definition is the showstopper for unions. If more technology means fewer employees, teachers will stuff their wooden shoes in the expansion slots. Economists like capital-for-labor substitutions, but educators revile them as threats to employment. And the specter of being replaced by a machine may have something to do with the rates of teacher use, which lag far behind the growth in installed base and connectivity.

Here's a telling example: A tech director for a Long Island school district spent the proceeds of a bond issue on hardware, software, and sustained professional development for teachers. Curious about utilization, he installed meters on the middle school classroom desktops. A month later, he had determined that the fans were blowing 100 percent of the time but the average desktop ran executable code seven minutes a month.

Teachers determine more of the reality of what students experience in classrooms than do any other class of actors in school policy. We believe them to be “closest to the children”; we grant them tenure; and there are simply too many teachers to be closely supervised, even if that were desirable. As a result, teachers are the final arbiters of all classroom improvement policies, including learning technology (in the schools). Learning technology must make them more successful, or it will not be used.

Before his death, Al Shanker, longtime president of the American Federation of Teachers, said that teachers had completed their cash-in-fist, wages-and-working conditions agenda. He said they and we should turn next to professional questions such as accountability and technology. Shanker knew that technology was coming, just as he knew accountability was coming. He used to say he didn't want fewer adults in the classroom; he wanted the adults in the classroom doing different things. His vision was a classroom in which technology surrounded the children and the teacher and freed the teacher to do the things that people do best. Digital tools should be used to do things that RLHBs (Real Live Human Beings): (a) should not bother with, such as keep records; (b) do not want to do, such as drill children; (c) cannot do, such as have infinite patience; or (d) do not do reliably, such as treat children of different races and backgrounds as though they all can learn.

Digital techniques change diagnosis, storage, and presentation in ways that teachers charged with the care of 22 children can never approach (see Figure 4). For example, is it really possible to "individualize" instruction in a class of 22 students whose grade-level performance spans three years? Think of the practical, logistic limits of the teacher's life.

Far from being a job threat, technology should be a job enhancement. Rather than substituting technology for teachers, this supplements carbon-based teaching with silicon-based learning. Learning technology is, or should be, an additive strategy.

School administration is consumed with the management of teachers and teaching (consider “standards-based instruction”) when it might be centered on the management of learning. Policy makers, administrators, and supervisors go to great lengths to influence the teachers' work defined only as full frontal instruction. But recall John Dewey's stinging conclusion in How We Learn: “We practically never teach anything by direct instruction but rather by the creation of settings.” Mark Twain said the same thing differently: “Never confuse schooling and education.” With digital, we now have what neither Twain nor Dewey had — the ability to construct micro worlds and use them to advance learning.

School administration is consumed with the management of teachers and teaching when it might be centered on the management of learning.

As long as we could believe that education came (only) from schools, then it was acceptable for the United States to have a Department of Education that functioned as a Department of Schooling. Australian governments have departments of “School Education” located in Ministries of Education that include portfolios for culture, sport, telecommunications, and schooling. There will be more learning in this country when we attend simultaneously to what Lawrence A. Cremin described as “the other educators.” That will mean adding learning to teaching; parents to teachers; homes to schools; and critically, technology to instruction and learning.

Dale Mann is managing director of Interactive, Inc., and, for 30 years was a professor at Teachers College, Columbia University. He lives in Huntington, New York.