Do economics majors need to learn Excel?

My first job after finishing my undergraduate degree in economics involved using Lotus 1-2-3 - the first "killer app" spreadsheet program - to create graphs. I'd never been taught to use a spreadsheet, but I worked it out.

Fast forward a couple of decades. Spreadsheets are ubiquitious in the workplace. When a new research assistant joins the Bank of Canada, their first job - like mine - is crunching data with a spreadsheet. Yet, at some universities, it is possible to graduate with an economics degree and never learn how to use Excel.

I'm deeply cynical about the usefulness of training. Having wasted hours mastering Wordperfect and other obsolete programs, I know that specific human capital depreciates rapidly, and is often a poor investment.

Yet spreadsheets are used in so many applications that it is worth getting some basic knowledge of how they work. Spreadsheets, moreover, provide a crucial conceptual basis for linear algebra and econometrics. A matrix is basically a spreadsheet, a vector is basically a column, an observation is a row. Understanding data in the abstract is difficult, picturing it as a spreadsheet is far more straightforward.

So why aren't our students learning how to use spreadsheets?

Spreadsheets seem to be surrounded by a Someone Else's Problem field. High school teachers figure students will learn to use them in university; university professors assume students must have been taught to how use spreadsheets in high school. Theorists and people who teach field courses leave spreadsheet instruction to the business stats instructors - and business statistics profs sometimes do provide spreadsheet tutelage, especially those teaching in business schools. But not every student takes a spreadsheet oriented business stats course.

I don't think that economics degrees should be just like business degrees - I like the fact that economics has strong theoretical underpinnings, that conceptual understanding and rigor is emphasized. Yet at the same time, don't we have a responsibility to ensure that our students graduate with some basic office skills?

Comments

[Determinant]

Just demand it in the course and have them read the manual. That's how I learned Excel and MATLAB. Though when a final-year prof wanted a complicated project in Visual C++ I thought she was *&^%$&*%.

On the subject of durable training investments, I am currently taking online French lessons in order to qualify for the BBB standard for the Public Service. BBB French is attached to pretty much everything that moves is in the Public Service. If you want to get into management, CBC is required. (The difference between B and C Speaking is that C-level can reply to a subjunctive question in the subjunctive.)

Three cheers for the Internet for making the excellent second-language French industry in Ottawa available to the rest of the country.

Just to warm Jacques' heart, I am improving the bilingualism statistics of this very English county in Central Ontario. :) And reading La Presse articles to improve my comprehension.

I will end this on a bilingual note: I move that Thomas Mulcair, the "Grizzly Bear", be known to one and all, especially to editorial cartoonists, as "L'ours brun d'Outremont."

[Linda]

Frances, where in the program would you propose spreadsheets be taught? At UVic we have a three-course stats/econometrics sequence for our majors; I cannot swear that they see EXCEL in any of those courses, but I am willing to bet they learn enough to at least be able to read the manual.

[Frances Woolley]

Linda - I'd teach it right alongside calculation of mean, median, etc in the intro to stats course. There is no point in teaching students (X'X)-1X'Y if students don't really understand what a matrix or a vector or a variable or an observation is.

I'm doing spreadsheets right now in my public finance course, though.

[Greg Tkacz]

Excel hasn't changed much since I first started using it back in '92, so unlike Lotus it appears that Excel skills have a long half-life. I introduce it to my students (especially Macro and Econometrics) whenever I download data live in class, and any respectable data source -- e.g. CANSIM, St. Louis Fed, OECD -- allows you to save the data in .XLS format, so you can quickly show students how to compute growth rates and prepare charts in less than 60 seconds. Those who look-up from their laptops long enough to see what I'm doing appear to be impressed.

[DavidN]

Use Apple Numbers instead, better looking graphs1

[Alex Tabarrok]

In my principles textbook with Tyler Cowen, Modern Principles, we have several optional Excel sections showing (for example) in Micro how to use Excel to solve a price discrimination problem and in Macro how to use Excel to show the Solow model.

[Jim Sentance]

We make our majors take a first year course our Computer Science department puts on, which among other things introduces them to excel.

[Ryan]

In my opinion, anyone with a mind capable of understanding linear algebra has all the building blocks required to figure out Excel. Not everything should be part of the core curriculum. Students should have to figure some stuff out. Figuring out which software applications and programming languages will be needed by "tomorrow's economists" is a losing battle, anyway.

With so many econ graduates unable to explain core concepts like comparative advantage and the non-neutrality of money, I'd say we should invest every ounce of effort into pure theory and let students pick up the peripheral skills as they go.

[Frances Woolley]

Ryan: "With so many econ graduates unable to explain core concepts like comparative advantage and the non-neutrality of money, I'd say we should invest every ounce of effort into pure theory"

Not sure if this is a serious response or subtly brilliant satire. How much of a typical pure theory course is devoted to getting students to explain concepts?

The question is not whether or not students *can* pick up Excel. Generally speaking, they can. The question is whether or not learning Excel or some other spreadsheet program is beneficial, either for building understanding of economic/econometric concepts, or finding gainful employment.

[whitfit]

I disagree with Ryan. I am not an economics prof, but did do an undergrad and masters. Although we used Stata in my graduate econometrics courses, I'm sure I was introduced to Excel somewhere along the way.

I was never very good at econometrics, doing better in theory/calculus/game theory etc..., but I have to say, being able to understand how a column or series of data can be manipulated is very valuable to understanding how statistics work. Comparative advantage and the non-neutrality of money are all very well and good, but they are theoretical concepts. In order to see how they apply, being able to use a spreadsheet program is useful. You could maybe get those students to set up a couple of examples of comparative advantage using a spreadsheet, and maybe after playing with it, they'd understand better!

And with respect to software applications and programming languages becoming obsolete, I don't think that it matters so much for these statistical/data tools. I can't imagine the spreadsheet of the future is going to be so different from the spreadsheet of today. You enter data, a formula for manipulating it, and get your results. Sure, some of the tools or commands might change (especially for graphing), but a sum/median/mean of a series is not going to be done a whole lot differently as far as I can imagine. If the tool changes slightly, knowing how a spreadsheet works will still be helpful. Part of what you do is enter formulas with respect to a data series. That is not about software as it is about being able to think how that data translates into a formula for getting to a certain goal.

But then again, I am a big fan of spreadsheets. The other day I showed my wife my spreadsheet for our lifetime finances and she thought I might be a little bit crazy ;)

[Neil]

The problem should be dealt with squarely in the high school realm. Even people who don't go on to university should be able to use excel, as it's the most versatile business tool out there, and is extremely inexpensive. I spent a couple days recently cleaning up a spreadsheet tracking construction change orders on a hospital reno (a three year project with around 1200 changes)...the fees didn't add up because the admin that maintains it was doing all the math with a calculator and typing in the results.

This would also conveniently solve the problem of universities - which are clearly intended to teach theory and expand the minds of their students - from getting bogged down in providing job training.

That said, in the absense of adequate high schooling, you should make sure that a supplementary course is available to students for a low price, and highly recommend that anyone without an intermediate understanding of the program should take it. Reading the manual is a start, but there's a lot of features that save loads of time on data analysis (I'm thinking of pivot tables as a prime example) that no one would ever think to look for.

[Frances Woolley]

Neil: "The problem should be dealt with squarely in the high school realm"

I was actually really surprised that, given the huge amount of time that the high schools here spend teaching students how to use graphics calculators, spreadsheets are not used much, if at all, except in the grade 12 data management course (though admittedly I haven't really studied the Ontario high school curriculum). It is such a basic workplace skill.

It's interesting that you mention pivot tables. When I was asking a econ forecaster to suggest videos for the MumblingProfessor youtube series, he mentioned pivot tables. Snag is, of course, that I'd have to figure out how to use them first!

[Andrew F]

My university taught me how to use Matlab, R, etc. but never excel. In fact, I was banned from taking a course where I might learn such skills for credit as a CS major. I knew some basic stuff from high school. I never got all that involved in excel's fancier functionality until I started work as an analyst a few years ago. Years and years grinding away with spreadsheets and databases. I actually kind of like it--it's like Sudoku that isn't utterly futile.

[David Giles]

Frances, Linda: Yes, our students should "know their way around" Excel. In the undergrad. econometrics courses at UVic. they meet it when accessing/importing data into other packages. I guess they have enough intuition (or something) to pick up all they need, because whenever I sit down with a student and they are using Excel, they are WAY MORE competent than I am.

Of course, no one in their right mind would use Excel for any serious statistical analysis - its shortcomings are too well known & documented. I discussed some time ago in a blog post at http://davegiles.blogspot.ca/2011/08/beware-of-econometricians-bearing.html

[Frances Woolley]

David - I was hoping you'd join in the discussion, knowing that this was something of a pet peeve of yours.

Who do you think has a responsibility to teach students the skills Greg Tkacz describes - downloading data, showing growth rates, etc - people who teach macro or field courses, or people who teach statistics/econometrics?

If the former, this raises fundamental questions about the nature of econometric teaching - what is econometrics/statistics for? Is it about quantitative methods in general, or is it about a specific set of skills related to a particular type of analysis? How do we learn mathematics? In particular, what is the relationship between numeracy and higher level mathematical skills?

For example, you might assume that your students know what data is. Perhaps they do. But I'm finding in my fourth year course that when I ask students to "go out and download some microdata" a significant minority really struggle to understand what micro data, as opposed to time series data, is (I do explain, or at least, try to explain, what I mean, but it's not a concept that people just intuit right away). Playing around with a spreadsheet might help - at least, that's what I'm going to try next year as an introductory exercise, just to get people to start thinking about data and coding issues, so it isn't such a huge barrier to them going forward in terms of actually doing some fun econometric work.

If you're unconvinced, here's something to try out on your own upper-year undergrads (or even MA students): flash a couple of types of data on the screen, e.g. employment numbers from the labour force survey, and data on NHL players. Ask, if they were estimating the relationship between GDP and employment, where would the standard error be coming from? Ask if they were estimating the relationship between NHL player height and NHL player salary, where would the standard error be coming from? These are profound questions, and really crucial ones for understanding econometric analysis - so it would be interesting to see how your students would do.

[Bruce]

"Spreadsheets are ubiquitious in the workplace. When a new research assistant joins the Bank of Canada, their first job - like mine - is crunching data with a spreadsheet."

That's true of the regular commercial banks too. I completed three university degrees and spent almost all of my time in MS Word. My Excel skills were "moderately" good as a result, but it's only in the workplace that I've really gotten good at it. [I also paid about $500 to take a course in Excel at a private career college here in Toronto and that was extremely valuable]. I've heard it said that university is actually all about getting your first job (i.e. after 2-3 years in the workplace, your workplace performance overshadows education). In that case, if putting energy into technical skills helps with the technical skills then it is a great idea.

[Ryan]

Frances - Wasn't satire at all. Peter Boettke has written that consistently applying the most basic concepts of economics is one of the hardest things an economist can do. I think abstract concepts, i.e. the real foundation of economics, are where most teaching effort should be placed. When I was in school, the ability to apply pure theory was the difference between a good student and a great one.

I don't know, I guess it comes down to how you view education. If the purpose of an education in Economics is to prepare oneself for the job market, then sure teach Excel, and SAS, and optimization techniques, and applied statistics and... suddenly you're looking at a degree in Statistics, not Economics at all.

But if the purpose of an education in Economics is to learn economics, then I say spend more time on the core theoretical foundations. Just my opinion.

[Frances Woolley]

Bruce - interesting. At what point will a commercial bank looking for new, retail level hires say "I'd rather hire someone from a commercial college who can use Excel properly than someone with a BA in Econ who is working it out by using the manual"?

If this happens, universities should be afraid. Very afraid.

[Frances Woolley]

Andrew F - how do you respond to David Giles on econometrics?

[Frances Woolley]

Ryan: "When I was in school, the ability to apply pure theory was the difference between a good student and a great one."

This is what I tell my fourth year students, after they have spent a couple of months in pure agony working out how to download and recode data: the difference between the A papers and the B papers is the theory.

What I do in the fourth year course, though, is make everyone draw indifference curves and budget constraints or something similar (not every problem can be reduced to indifference curves and budget constraints). Then say "what will cause the slope of the budget constraint to change in your particular situation?" "What kinds of factors change preferences?" This tells the student what to include as explanatory variables in their econometric model. E.g. one student is looking at marijuana use, so the question is "what kinds of factors change the relative price of marijuana and alcohol?" (e.g. provincial differences in legal regimes, the amount of marijuana grown locally etc).

What is interesting, though, is that it isn't always obvious to students that this exercise - which looks so simple, and which they appear to find somewhat useful - is in any way related to the theory of consumer demand, or the comparative statics, that they're tearing their hair out over in their advanced micro course.

[Mickey]

Interesting question. I come at it from the perspective of an employer (in a quantitative finance field) where the use of Excel is constant and required.

To succeed in finance (fyi, I have a background in Economics) you do need spreadsheet skills - but, before you need spreadsheet skills you need the strong theoretical underpinnings that have been discussed. People who can blast numbers out of a spreadsheet are a dime a dozen. What is more rare is the person who can make sense of these numbers.

The time investment on learning Excel is very modest compared to the time required to understand the stories data tell. As an employer I would much, much rather have the smart, lateral thinking graduate than the graduate proficient in IT. The latter can be learned in the workplace very quickly. The former can also be learned, but much less quickly.

So, if I had a vote I would suggest universities not waste valuable time in teaching specific Excel techniques (unless it is integrated into the real teaching - like the way research skills aren't necessarily specifically taught, but can be a derivative benefit from being tasked with writing a paper).

[Bob Smith]

"The problem should be dealt with squarely in the high school realm"

Sure, high school graduates can't use excel. But they've got great self-esteem, so what more do we want?

[Frances Woolley]

Mickey - this raises an interesting question. When 10% or 20% of the population was graduating from university then, yes, people with university degrees would be primarily going into the kind of jobs that you describe - ones that require smart, lateral-thinking graduates.

When 40 or 60 percent of the population is graduating from university, both the nature of the student body, and the kinds of jobs that students can expect to get upon graduation, changes.

Shouldn't universities respond to this massive expansion in undergraduate enrolment by changing - at least somewhat - what is taught in the classroom?

[Andrew F]

Frances, I agree that excel is less than ideal for heavy-duty statistical analysis. There are many packages that do a better job, but they are often expensive and your employer may not be willing to plump for a $5k software package on your say-so. So excel is useful for quick-and-dirty analysis, or for managing data sets that are used by other applications (you can do transformations to data easily in excel, and it is a better csv editor than most).

I mostly use excel for reporting and any back-of-the napkin brainstorming I need to do. I use SQL for data processing whenever possible (often dealing with data sets with 100 million records, which make excel cry).

[Andrew F]

I think it also worth pointing out that as many disciplines become more specialized, it takes longer for students to learn the theory required to reach the frontier. And knowing how to use software tools can be just as useful for an academic as for someone working in the private sector.

I think it is worthwhile for universities to spend time teaching how to use tools. Preferably it would not be too application-specific, but teaching how to use a class of tools, like databases. Perhaps universities could partner with colleges to offer college courses on campus.

[richard]

"Ryan: "When I was in school, the ability to apply pure theory was the difference between a good student and a great one.""

Perhaps learning to evaluate the theory using real data is more important. At that point a lot of the garbage can and should be thrown out. Instead, it seems that most times the data get thrown out, because they inconveniently invalidate the theory.

I went to uni in the stone age, before pcs were on the scene, and when engineering students carried slide rules around like swords. So I ended up learning computer programs on my own. I'd agree with a previous poster that Excel, given its track history, should be taught in high school. At the very least, it should be mandatory in first year uni, but this can be done easily by integration into existing courses (there are enough on-line resources that the loss of instruction time will be minimal).

"(The difference between B and C Speaking is that C-level can reply to a subjunctive question in the subjunctive.)"

Sorry, Determinant, but these exams are now often done by phone, and the move from B to C depends as much upon having a well-rested and good-humoured interviewer as it does the use of subjunctive.

[Ryan]

Richard - Good point. I should be clear that I don't actually object to the use of applied statistics or the learning of Excel. I guess my point is, the closer an Economics curriculum gets to teaching real-world job skills, the more it starts to look like a degree in Statistics.

Put another way, for those who favor teaching Excel as part of the econ curriculum, what in your opinion is the key difference between a degree in Economics and one in Statistics, and how might the curriculum best be varied to accommodate your point of view?

[Frances Woolley]

Andrew F "There are many packages that do a better job, but they are often expensive and your employer may not be willing to plump for a $5k software package on your say-so."

I'm wondering if the wave of the future may well be Excel for basic data manipulation and graphics, then R for serious heavy duty analysis. A cheap solution to please both the budget conscious employer and the analyst looking for serious power.

[Frances Woolley]

Ryan: "I guess my point is, the closer an Economics curriculum gets to teaching real-world job skills, the more it starts to look like a degree in Statistics."

Wow. And I thought I was cynical and pessimistic.

[CBBB]

Why don't they teach SAS or something? I see tonnes of jobs that are relatively entry-level but they want a solid amount of SAS experience. SAS is expensive and you really can't afford to learn that on your own. I graduated with a Math degree a few years ago and it's been totally useless but I think it would be a lot more marketable if the school had included more courses that focused on learning and using packages like Matlab and SAS (not R - HR reps don't know what that is so it's not useful for your resume to learn R).

[richard]

"I guess my point is, the closer an Economics curriculum gets to teaching real-world job skills, the more it starts to look like a degree in Statistics."

Yes, Ryan, but Stats has more value than Economics - mainly because too many economists favour theory over data.

[Joel W]

I think Excel experience is a very useful skill for undergraduates to have. I use Excel for personal budgeting at home, and as an economist virtually every week at work.

Curriculum mapping can probably ensure that Excel skills are learned along the way, rather than mandating all econ students to take Intro to Computers.

In undergrad at UBC I learned Excel via Intro to Computers (Comp Sci 100), Environmental Economics, and my 4th year Applied Economics class (half of which was all time series econometrics).

I think Intro to Economic Statistics (or whatever it is called at different econ depts) is probably the best place to use it. In my undergrad we used SHAZAM instead, which was utterly useless for understanding things like mean, median, random walks, etc. IMO Excel is much better for that.

Environmental Econ is also a good place to include Excel in the curriculum. It is great for teaching discounting, and for graphing trends in environmental variables. When I taught a 2nd year "Growth and Environmantal Quality" course, I had an assignment where students needed to download environmental data and then produce time series graphs in Excel, and then write a short essay interpreting the graphs.

[Frances Woolley]

CBBB: "I graduated with a Math degree a few years ago and it's been totally useless but I think it would be a lot more marketable if the school had included more courses that focused on learning and using packages like Matlab and SAS"

Joel W: "In my undergrad we used SHAZAM instead, which was utterly useless for understanding things like mean, median, random walks, etc."

I hear you. I don't know how to improve students' lives though.

[Ryan]

Frances - I'm not trying to be pessimistic. Think about how your question might apply to a Philosophy department. "Should we teach Philosophy students how to use Microsoft Word?" You very well could, but then you're not really teaching philosophy anymore.

We all know that Philosophy isn't the best course of study when it comes to "getting a real-world job," but for philosophy students, that's not really the point.

It's not pessimism to acknowledge that Economics is a social science like sociology, history, philosophy, etc. The difference between Economics and Statistics is theory. Strip away the theory, and you've just got a lot of optimization and regression techniques. It's not cynical to say so. I studied Economics and I'm glad I did - because I wanted to learn about economic theory. If what I wanted to know was spreadsheets, database manipulation, and applied statistics, I would have minored in Math.

Oh wait - I did! :)

[RebelEconomist]

Yes, yes, yes! Frankly, such general skills as maths, statistics and computing will be more useful to most students than the economics, because they seem to involve the type of thinking that is easier to pick up when you are younger. Any economics department that really cares for its students should make those subjects a substantial part of the course. In fact, if an economics course has a substantial finance content, you could even consider introducing the students to VBA. I dare say that many if not most economics students will go on to be used as general numerate analysts, and the ability to crunch numbers and plot the results quickly will give them a flying start. I learned Microfit in my economics MSc in the early 1990s, but I never used it working in a central bank, because I never did a research job (and even there, they tended to use RATS). Such specialised tools come and go and are only available locally, whereas Excel has been around years and is used in most numerate jobs. I used Excel frequently, and still do.

[Frances Woolley]

Ryan: " "Should we teach Philosophy students how to use Microsoft Word?" You very well could, but then you're not really teaching philosophy anymore."

One highly valuable skill that philosophy teaches is how to write a coherently structured essay.

Writing a coherently structured essay is central to learning and applying philosophical reasoning. In the same way, certain analytical tools are central to learning and applying economic reasoning. Philosophy profs don't say "I'm not going to worry about whether or not your essay has paragraphs, you can learn that on your own." There are parallel skills that econ profs should expect of their students. In the past I wouldn't have said that Excel was one of those skills, but now that students are spending so much time on stats/econometrics, I'm starting to think that perhaps it is.

[Ryan]

Wow. I think you just trumped my argument. Haha! :) Kudos.

[Michael Bishop]

If I ran a department, there would be a lot of excel, though people could place out of it. Hell, I'd require a certain amount for the entire university if I had my way.

[Greg Tkacz]

I was gobsmacked when I became a full-time academic to discover that many students had never even heard of CANSIM, so by extension I saw it my duty to show them how to find, retrieve and manipulate some basic economic data. It's one thing, for example, to draw a generic business cycle on the board, but it's more revealing to students to show them how the economy has actually evolved over the past few years. This is where Excel enters the fray, as most databases allow you to save your data in Excel format, and most statistical packages are able to read Excel files. It's become the useful (or indispensable?) data-manipulation intermediary that students need to be exposed to if they ever have a hope of working in their field upon graduation. Whether we like to admit it or not, Economics is a data-driven subject (as evidenced by the 40+million CANSIM series), so a basic knowledge of tools that allow you to retrieve and manipulate data is crucial.

[Frances Woolley]

Ryan "I think you just trumped my argument." Clearly those undergrad philosophy courses coming into play!

Greg - would you like to join the Economist Party's radical wing - the Data Liberation Front?

[Determinant]

I thought the DLF disbanded; Statscan made CANSIM free and the paywall came down on New Year's Day.

richard:

I have a friend who complained he got bumped down from a C to a B on Oral since he didn't reply in the subjunctive to a subjunctive question.

I only aspire to a B right now since BBB is what is attached to what I am interested in.

[Scott P Bacon]

This article is fascinating... I love any environment where economics and philosophy collide...

Neil: "The problem should be dealt with squarely in the high school realm"

I agree halfheartedly. Without the basic understanding of excel I obtained in high school, I would not have been able to maintain my small business. As a result I would not have been able to attend University. But my skill in excel was greatly complimented by my training in statistics (Stats 2, Econ 2201/2, etc), so to say that high school training provides the best background, or even the best starting point for a foundation in excel, I think, is misleading.

@Frances "spreadsheets are not used much, if at all, except in the grade 12 data management course"

I took the grade 12 data management course, and from what I can remember, it was almost exactly like my Econ 2201 (Stats 1) course; albeit Econ 2201 had additional material by the time we got to the ANOVA.

Moreover, I think you are incorrect in assuming that high school students are not exposed to excel in any other courses. I distinctly remember: 1) my grade 9 introduction to business class where the whole Microsoft office suite was covered in some detail, 2) grade 11 computer engineering where Visual basics, Excel, PLC programming and other material was covered, and 3) grade 11/grade 12 accounting which (perhaps obviously) relied primarily on the use of excel!

However I think your inquiry leaves us with at least one remaining question, briefly touched upon above: should students who choose a more mathematically oriented field of study, say economics, be required to take on a more specialized training in excel to further their understanding of the theoretical concepts they will become exposed too? I think the answer is a definitive yes.

(P1.) If (economics) students are required to recollect and review mathematical material, say like introductory calculus and linear algebra (previously Math 1009 and Math 1119 at Carleton), and
(P2.) these classes include basically the same material covered in their high school equivalents, save the addition of a few new topics, then
(C.) I do not see why a more advanced/specialized training in excel is not also warranted.

If the above holds, then I think your (Frances) suggestion of teaching excel "right alongside [the] calculation of mean, median, etc in the intro to stats course" to be appropriate. As I pointed out above, there was a great deal of overlap between my experience in grade 12 data management and that of Econ 2201; at least as similar as any of my university introductory math classes and their high school equivalents/prerequisites.

[Kailer Mullet]

I think it's important to distinguish what's useful for students to know and what's useful for students to be taught. I don't know anybody who can learn a programming language by attending a lecture. Why do you think so many silicon valley moguls are college drop outs? There's only one way to learn programming, and that is by having a particular problem you want to solve, then looking on the internet or in help files to learn how to solve that particular problem.

That said, if you are going to teach a course on programming I think excel/VBA is the way to go. VBA especially, as it is an easy language to learn and is eminently practical.

[Jciconsult]

Economic students should be required to learn Excel (particularly they should be able to make pivot tables dance). They should also be required to learn R. There is no excuse for using weaker software systems.

[Frances Woolley]

Kailer - "I think it's important to distinguish what's useful for students to know and what's useful for students to be taught."

That's an excellent point. At the same time, I don't know if those of us who starting teaching B.P. (before powerpoint) always think about the potential use of Excel in teaching.

E.g. today I was teaching capital cost allowances (corporate taxation) and I did something new - pulled up a spreadsheet, used it to do the CCA calculations, and then copied and pasted, so people could see how CCAs evolve over time, approaching but never reaching zero. More excitement and understanding than I've ever got in the CCA stuff before (course that's a very low bar, and say more about how difficult it is to follow my mathematical calculations on the board than anything else, but still...)

[Alex Godofsky]

Excel is really easy to us (by design!). For a lot of the jobs we're talking about, if you can't pick up the skills yourself the employer doesn't want you anyway.

[excel classes]

Hmmmmp! Get out there and sell. You don't need Economics, it does help but is hardly essential. If you are struggling with it drop it, and start looking at how to make money in your spare time. What special fields do you have knowledge about, what interests you, what hobbies, get out there and start buying and selling. I bought a laptop for £100 over the weekend, HP 6735s, if I wanted to i could resell it on for £180, by buying a new battery for £30. There is a guy locally who buys and sells Laptops and makes a nice sideline, just because he knows what goes wrong, and that he can unscrew the back and put it right.

[stickman]

Bruce and Joel W nail it.

Of the career possibilities available to econ majors, a vast majority will involve substantial time staring at Excel spreadsheets... whether you are a management consultant or collecting data for the UN. (Having worked outside of academia for several years before returning to take my PhD, I can personally testify to this.) If we aren't training our students to effectively use software that will be integral to their professional lives -- at least the overwhelming majority that aren't going to become economic professors -- then we aren't doing a proper job as educators.

A minor, but not entirely trivial matter is the fact that most established datasets come in Excel form anyway. Anyone tried to download data from a government website or NGO portal recently? Copying across to Stata or whatever might be simple, but some familiarity with Excel is important if you want to run basic operations within these portals themselves.

2) Courses in environmental and resource economics are great places to start with Excel, due to the reasons that Joel W mentions... discounting, resource flows, inter-temporal optimization (e.g. the "Solver" add-in), etc. My own resource prof insisted on using Excel for the exact reasons that I have outlined above.

I would say that the exact same is true for much financial economics as well.

[Tom]

Econ 001 or Econ 101 should require beginning Excel skills. (My High School college counselor counseled against learning Typing, but it was one of the best, most useful courses I ever took.)

Students don't need to become experts (not VBA, nor R nor SAS -- tho these are great), but should be required to be able to open a spreadsheet, and download some data, and create a graph from it.

The ability to use data, and verify (falsify?!) some theory is, in practice, likely to be one of the most important abilities most students learn.

PS: Francis, it was Visicalc for the Apple II that was the first real killer app, and I remember 1980 just before IBM came out with a PC that the head of Profit and Budget Analysis took his Apple II to work to use Visicalc for his budgets. Had IBM waited another year or so, Apples might have been entrenched enough to allow more Lisas to be sold (and delay Mac development?)... long ago and far away in Sili Valli. Lotus 1-2-3 in 640k RAM was, relatively, FANTASTIC -- single most important program in PC history, still.

[Frances Woolley]

Stickman "A minor, but not entirely trivial matter is the fact that most established datasets come in Excel form anyway."

Agreed, and you're right, this is not trivial. Our students learn Shazam in econometrics. The only practical way to download data and input it into Shazam, as far as I can tell, is to download into Excel, save as a .csv or similar, and then upload it that way.

Tom - I stand corrected! On Lotus 1-2-3 - yes, you're right, it was fantastic. There is almost nothing I do in Excel right now that I couldn't do as easily in Lotus 1-2-3. Some things in Excel are harder, because the standard formats for graphs don't always work very well in mono-colour.

[Alex Plante]

Definitely yes. I'm an engineer, and you'd be surprised how many engineering calculations are performed using Excel. I learned fist Lotus 123 then Excel "sur le tas" as we say in French, and finally took a night course in Excel. Many of my colleagues made fun of me, as if I were taking a course in basket weaving, but it was actually one of the most useful courses I've ever taken.