Forgot your password?
typodupeerror
The Internet Education

Students From States With Faster Internet Tend To Have Higher Test Scores 175

Posted by Soulskill
from the in-before-correlation-!=-causation dept.
An anonymous reader sends word of correlation found between higher internet speeds and higher test scores. Quoting: The numbers—first crunched by the Internet provider comparison site HSI — show a distinct trend between faster Internet and higher ACT test scores. On the high end, Massachusetts scores big with an average Internet speed of 13.1Mbps, and an average ACT test score of 24.1. Mississippi, on the other hand, has an average speed of just 7.6Mbps and an average score of 18.9.

In between those two states, the other 48 fall in a positive correlation that, while not perfect, is quite undeniable. According to HSI's Edwin Ivanauskas, the correlation is stronger than that between household income and test scores, which have long been considered to be firmly connected to each other. The ACT scores were gathered from ACT.org, which has the official rankings and averages for the 2013 test, and the speed ratings were taken from Internet analytics firm Akamai's latest report.
This discussion has been archived. No new comments can be posted.

Students From States With Faster Internet Tend To Have Higher Test Scores

Comments Filter:
  • by wbr1 (2538558) on Friday August 22, 2014 @01:07PM (#47730729)
    Faster internet access means faster internet search results when cheating. Therefore the internet should be banned. /s
  • Cheating (Score:2, Insightful)

    by Anonymous Coward

    Cheating over dial-up sucks!

  • by mcl630 (1839996) on Friday August 22, 2014 @01:07PM (#47730741)

    The usual /. refrain of correlation is not causation definately applies here. Mississippi had low test scores long before broadband Internet came along.

    • by PPH (736903) on Friday August 22, 2014 @01:10PM (#47730783)

      Yep.

      Low test scores correlate with low income. Low income correlates with not affording premium services.

      • by cp5i6 (544080)
        No,

        low income and low test scores have been shown to be a strong correlation with some studies showing a causative factor.

        low income and not being able to afford premium services is a definitive causality of one to the other


        The correlation is the ACT test scores to premium internet services unless as someone pointed out, the internet was being used to cheat on the ACTs, in which case it becomes causative.
        • Re: (Score:3, Informative)

          by Pfhorrest (545131)

          Correlation does not imply causation, but causation does imply correlation.

          If A causes B, then A will also correlate with B. It's only the reverse that's false. (A correlating with B doesn't mean A causes B).

          • by i kan reed (749298) on Friday August 22, 2014 @02:39PM (#47731681) Homepage Journal

            No, causation can actually have a negative correlation, surprisingly.

            It's relatively easy to generate cases where X can be caused by 2 different things, but the majority case simply has a stronger effect, if you don't account for mediating factors.

            Let's take a case where everything is controlled by people so cause can be quite directly ascribed.
            Applying for US citizenship causes US citizenship.
            But among residents of the US, those applying for US citizenship are far less likely to end up being citizens than those who don't.

          • by cp5i6 (544080)
            without going into stats, the short answer is no, that's not right.
          • by cp5i6 (544080)
            acutally i like this example if you were looking at a case where

            a bully beats up a kid and takes his money
            the mom feels bad and gives more money back.

            the causation case is that when the bully beats up the kid, he gets more money.

            without knowing what the mom is doing there is no real correlation (ie, you can't say that everytime a kid gets beaten, a kid gets more money), only that there is causation of one to the other. If however, the mom is known, then you can say that everything this kid gets
        • by Anonymous Coward

          It's also worth noting that ACT scores may not be the best metric for success.

          In some states it's mandatory for high school students to take the ACT, which may lead to lower state averages (technically those lower averages are more accurate for the whole population, but still don't compare directly with states where only the best students take the test).

          Moreover, SAT is more popular on the coasts. Many of my high-achieving friends (think West Coast Ivy League) never even took the ACT. With that in mind, how

      • That's true but the article (and even the summary) says that the correlation with internet speed is stronger than with income. So there may be more to it than just rich people can afford premium services.

        Maybe families that value education more strongly are more likely to get broadband, or maybe there's is actually some causation.

      • Re: (Score:3, Insightful)

        by DigiShaman (671371)

        Low income means that parent[s] need to work longer hours and spend less time with their children. Family time has a profound impact on education.

      • Low test scores correlate with low income. Low income correlates with not affording premium services.

        Beyond that, ISPs won't even offer high speed services in places with low population densities and low incomes- the kind of states that have low ACT scores.

        This study and this article are utterly stupid.

    • by mythosaz (572040)

      Mississippi, 42nd in literacy!

      Take that, Nevada, New Mexico, Georgia, New Jersey, Texas, Florida, New York, and California!

      [Immigrants aside, apparently, it's "Take that, Georgia!']

      • by mrchaotica (681592) * on Friday August 22, 2014 @01:39PM (#47731157)

        [Immigrants aside, apparently, it's "Take that, Georgia!']

        You think Georgia doesn't have a lot of immigrants? It's got a higher percentage than New Mexico (9.5% [migrationpolicy.org] vs 9.2% [migrationpolicy.org])!

      • by Richy_T (111409)

        Ratings are stupid. Of 50 states someone has to be #1 and someone has to be #50 and the slots in between* must be filled also. You could clone a billion Stephen Hawkings and replace the US population with them and somewhere would still be #50. It's just a trick used by politicians to extort more tax dollars from the populous.

        Sure, take a good measure and if you're far away from the mean in the low direction, you probably need to reassess your education program and if you're well above, you're doing somethin

        • by Richy_T (111409)

          I'll just add that TFA itself apparently doesn't fall for this (at least from what is in the summary)

    • by Tablizer (95088)

      Big Data is finally paying off: now one can sift jillions of bytes for hundreds coincidental correlations that used to take marketing departments and politicians several years and millions of dollars to concoct manually.

      The sweet smell of progress!

    • by NotDrWho (3543773) on Friday August 22, 2014 @02:01PM (#47731403)

      Who woulda done thunk that one of the poorest states in the Union would also have shitty broadband and shitty test scores? I bet they have crazy high infant mortality, shitty health in general, and a high per-capita crime rate too.

      Maybe the high infant mortality rate is what caused the shitty broadband.

      • by CrimsonAvenger (580665) on Friday August 22, 2014 @02:32PM (#47731639)

        I bet they have crazy high infant mortality, shitty health in general, and a high per-capita crime rate too.

        Per capita violent crime in MS ranks considerably below the national average (299 per 100K as opposed to 474 nationally (2006 figures)).

        Massachusetts, New York, and California all have violent crime rates rather higher than the national average....

        As does the District of Columbia (over three times the national average, nearly twice that of the highest State).

        • by jfengel (409917)

          Yeah, violent crime seems to go with density, rather than poverty. It's committed by the poor, but closely-packed poor rather than rural poor.

          DC is ALL urban, every single inch of it, so it's not really appropriate to compare it to a state. It's mid-pack compared to other cities of comparable size; it fell between Indianapolis and Miami on the 2012 list (http://www.fbi.gov/about-us/cjis/ucr/crime-in-the-u.s/2012), and near Toledo and Nashville.

          More urban states have higher violent crime rates, but it's cent

          • by Richy_T (111409)

            It's full of politicians and their hangers-on so it's going to suck in undesirable types like a sponge. I'm a "more guns, less crime" kind-of guy but even I think that that signal is swamped by other inputs in the case of DC

            • by jfengel (409917)

              DC's violent crime rates are largely about its poverty. Aside from the federal government, it has no real industry. It was heavily populated by poor blacks fleeing the South during the civil war, and during the next century-plus they were heavily discriminated against. There were few jobs for them except at the very bottom of service. While the place is on average pretty wealthy, its real population is quite poor.

              The real fail of the politicians was that for two centuries the federal government ran the plac

    • For fuck's sake... where in TFS or TFA was causation mentioned?
      Having a correlation between two data sets opens the door for more research on that matter. Nobody said "THIS is because of THAT", but rather "THIS and THAT scale similarly, hmm..." which is a totally different thing.

      So this finding might be further correlated with the following:

      - given the same amount of time spent on the Internet (say 1h/day as base value), the amount of information retrieved from the Internet scales with broadband speed.
      corre

    • by nucrash (549705)

      Let's see if I can account for also factors that are just as like culprits for lower test scores in Missisippi

      1. Unemployment rate, MA - 5.6 MS - 8.0
      2. Average Income
      3. Disposable Income
      4. Parents that actually care about their children's education
      5. Belief in a supreme being
      6. Racial diversity?
      7. Bio-luminescence
      8. More Shore line per state volume
      9. BP retarded the state with an oil spill
      10. I just put this one in here for giggles
      11. Mississippi is used to space out counting closer to a single second
      12.

    • by cmdr_tofu (826352)

      Agreed, there are huge societal and infrastructure differences between Masschussets and Missisippi. Mass might be one of the the most educated states in the union. Mississippi is not.

  • sorry (Score:5, Funny)

    by pezpunk (205653) on Friday August 22, 2014 @01:07PM (#47730745) Homepage
    • Re:sorry (Score:5, Informative)

      by pezpunk (205653) on Friday August 22, 2014 @01:11PM (#47730791) Homepage

      I mean somehow I doubt better access to broadband is going to solve Mississippi's education problems. I'm willing to bet Massachusetts kids were trouncing Mississippi kids long before the internet came along.

      • Re:sorry (Score:4, Informative)

        by mythosaz (572040) on Friday August 22, 2014 @01:24PM (#47730983)

        True.

        Back in '92, Mississippi was at the absolute bottom 50 of 50 in basic prose literacy. [Sates with high immigration have since pushed them up to #42..]

        In '92, Mass was in the middle of the pack, at about half as many illiterate residents.

        http://nces.ed.gov/naal/estima... [ed.gov]

      • It's almost like (barely) developing nations with pitifully low levels of human capital formation also have pitifully low levels of infrastructure investment. Funny that. It's probably their way of making our victory in the civil war seem completely pointless.
      • If they raise their test scores, we should give them faster internet as a reward.
      • by mlts (1038732)

        There is also the fact that Mississippi is a lot larger than Massachusetts. It is easy to build high quality Internet connections in a state that is small, with almost all of its population concentrated on the eastern side. A larger state with less population, and population that is more scattered, with the biggest town being about 1/20 the size of Boston makes it a lot more expensive to sling fiber and provide access to residents, especially in a state with such a relatively low population density.

        • by plover (150551)

          Interesting idea, but the data doesn't support it.

          While Massachusetts has 858 people per square mile, the population density of Minnesota, 68.1, is almost identical to Mississippi, with 63.7 people per square mile.

          U.S. Census data also shows a significantly higher percentage of residents with internet connectivity in both Minnesota and Massachusetts, and significantly lower percentage in Mississippi. (Sorry, the source, http://www.census.gov/prod/201... [census.gov], doesn't list the exact percentages, but I'm sure they

    • Why sorry? No-one's said there's causation between the two in either direction.

      • by pezpunk (205653)

        the sorry wasn't "sorry you're wrong" but rather "sorry or the obvious response, but it's applicable"

  • Yeah, before the Internet, everybody had the same test scores across the board.

    • Then it's time to look for a common cause behind low standardized test scores and unavailability of high-speed home Internet access.
      • by alen (225700)

        slow internet most likely means rural area where most people care more about sports than test scores

        only thing fast internet does is let you stream educational crap for kids since wikipedia doesn't need 13mbps

    • No, before the internet the rich didn't have faster internet access, but the better collection of books.

      If anything, internet access leveled the playfield a bit.

  • correlation does not equal causation

    • You can stop at "more $". That's the real reason why students in MA do better than students in MS.

      Not money that's used to buy kids iPads or Surfaces, mind you. Money that's spent to modernize schools built in the 1960s, or tear them down entirely and put up new ones. Money that's spent to pay teachers more, and attract better teaching talent. Money that's spent on the community and infrastructure to make teachers want to live there.

      Also, it's much more profitable for the private companies that "public" edu

  • This smacks of a 3rd factor that causes both faster internet and higher grades. It could be wealth, politics, or simply smarter people move to places with higher internet speeds.
  • Reading the linked article and looking at the graph I see that this is not even a good correlation never mind being causal.

    Want to have your kids do better? Pay attention to them. For the best results, homeschool.

    • For the best results, homeschool.

      There are plenty of countries where you can be thrown in prison for doing that to your kids. Germany is among them.

      • by bobbied (2522392)

        For the best results, homeschool.

        There are plenty of countries where you can be thrown in prison for doing that to your kids. Germany is among them.

        Which is a *good* reason not to live in Germany. Actually, most of Europe is not very willing to let parents homeschool. The Germans just take it really seriously.

        • The problem with homeschooling is that it's usually used not as a means to educate your kids better, it's most often used because you have a particular worldview you want them to be exposed to, to the exclusion of all others.

          • I agree. While I don't have data outside personal experience, and little of that, the people I know who home school their children do it because their child doesn't play well with others.

          • by bobbied (2522392)

            The problem with homeschooling is that it's usually used not as a means to educate your kids better, it's most often used because you have a particular worldview you want them to be exposed to, to the exclusion of all others.

            What world view parents can teach their children is THE problem with homeshooling? And that's a problem you want managed by your government? I don't.

            I don't think it's any of my business what the people across the street have for a world view, and they choose to homeschool? If the kids are learning how to read, write, do math, understand history and science what business is it of the government to impose a world view or a set of them to be taught? You think it's your business? How Orwellian of you.

            I can

        • by tepples (727027)

          Which is a *good* reason not to live in Germany.

          Good luck getting a pro-homeschool country to grant your family resident status.

    • For the best results, homeschool.

      If you think that having an 18 year old who reads very well and has never been on a date is "the best results", then sure.

      School is about more than books and tests.

      • by Richy_T (111409)

        You seem to know significantly less about homeschooling than you believe the gp does about schooling.

  • Faster internet also means a larger cable bill. Maybe we are seeing inflated test scores because the people with faster internet are the people who can afford it along with better schools, private tutors, school supplies etc etc.
  • by Atzanteol (99067) on Friday August 22, 2014 @01:28PM (#47731015) Homepage

    This is data mining. If you compare enough things you'll find strange correlations. There is little plausible reason to believe there is an actual causal relationship here.

    These are also "irrefutable correlations":

    US spending on science, space, and technology correlates with Suicides by hanging, strangulation and suffocation:
    http://www.tylervigen.com/view... [tylervigen.com]

    Number people who drowned by falling into a swimming-pool correlates with Number of films Nicolas Cage appeared in
    http://tylervigen.com/view_cor... [tylervigen.com]

    Per capita consumption of cheese (US) correlates with Number of people who died by becoming tangled in their bedsheets:
    http://tylervigen.com/view_cor... [tylervigen.com]

    • by Tablizer (95088)

      Per capita consumption of cheese (US) correlates with Number of people who died by becoming tangled in their bedsheets

      There may be a real cause/effect correlation there, but if so, I don't want to know what it is.

    • by tomhath (637240)
      Yup. And I also noticed they didn't print the standard deviation or p-value. Looking at the scatter chart there does appear to be a correlation, but it seems pretty weak to me.
    • by Anonymous Coward

      I will never eat cheese again.

    • Excellent link. Thanks.

  • by Anonymous Coward

    It's picking up other stuff omitted in the study.

  • between articles submitted with the term "correlation" in the summary, and with comments taking the article to task for being wrong about correlation implying causation.

    Nevermind that most of the articles make no such claim at all.

    But is it causal? Hmm..

  • If there was any data to suggest the ACT tests are statistically valid (they test the thing you think they test) or reliable (they would get the same result if you tested again) then the correlation may be a clue to something. However when the underlying test is neither valid nor reliable, the correlation it shows doesn't even show you there is correlation.

  • Not to mention other cross-level fallacies. See for example: Diez-Roux, A. V. (1998). Bringing context back into epidemiology: variables and fallacies in multilevel analysis [nih.gov]. American Journal of Public Health, 88(2):216--222.

    The upshot: Even if a causal relationship corresponds to the study's findings, causes of state-level rates of test achievement are fundamentally different things than causes of student-level rates of test achievement.

  • Rich kids tend to have higher test scores...
  • You can find all sorts of weird correlations [tylervigen.com] if you look for them but the mere existence of a correlation is meaningless by itself. In this case my first question would be about money. States with more money will be able to afford both faster internet and better schools. Other factors that need to be controlled for include population density, local industry, demographic makeup, etc to be able to put some meaning to this.

    Basically this is a meaningless correlation which provides no context to draw useful

  • Faster internet - more affluent - not black, etc etc etc

  • The more waves --electromagnetic, wires, fiber-- the larger the opportunity for so called 'rub off' of thise waves. Actually the better-performing students do not know anything more than the average student, nor have better insights, higher intelligence (technically though, yes they do), etc. It's just that they 'pick up' more of the accepted view of the current state of knowledge. Morphic resonance.

  • This just in!! Students with rich and strict parents have higher test scores ... This is News at NEVER ... DUH.
  • The states with highest internet speeds AND ACT scores are almost all Blue...
  • I'm looking for funding for a study to show a relationship between useless pseudoscience and states with high internet bandwidth per capita.

  • The expense of the housing in which a student lives has been a great predictor of success in schools. Expensive dwellings indicate an ability to spend and that means that service providers will invest more in hopes of capturing an account. Such homes can spend freely for expensive services such as HBO, Showtime, alarm systems and being the phone service as well. Whereas poor communities simply can not afford such things. It is not income levels that one measures. After all for the rich taxab
  • The ecological fallacy concerns making conclusions about individuals from aggregates (states).
    From Wikipedia,
    http://en.wikipedia.org/wiki/Ecological_fallacy
    "An ecological fallacy is a logical fallacy in the interpretation of statistical data where inferences about the nature of individuals are deduced from inference for the group to which those individuals belong. ... The four common statistical ecological fallacies are: confusion between ecological correlations and individual correlations, confusion betwee

  • by twistedcubic (577194) on Friday August 22, 2014 @11:05PM (#47734741)
    The students who take the ACT are not necessarily representative of the state's population. For example, I took the SAT, and not the ACT, because all the colleges I applied to accepted SAT results, but not all accepted ACT results. Students who were going to the local state school just took the ACT alone. I bet the reverse is true in different states. It doesn't take much thought to see these results are totally meaningless.

"Life, loathe it or ignore it, you can't like it." -- Marvin the paranoid android

Working...