Page 4 of 5 FirstFirst 12345 LastLast
Results 46 to 60 of 69

Thread: New Beach House/Electronic Voting

  1. #46
    Join Date
    Dec 2005
    Posts
    44

    Default

    Thanks Kim. That answers my question.

  2. #47
    Join Date
    Jan 2007
    Location
    Wayland, MA
    Posts
    235

    Default

    Quote Originally Posted by Kim Reichelt View Post
    As I saw this particular debate, somebody posted something on the Town Crier discussion board that Jeff D found unsubstantiated. To show that it was unsubstantiated, he presented data that suggested the opposite to be true. Didn't prove it, but suggested it. I'm sorry, Dave, but I fail to follow your logic in complaining that his suggestion isn't proven, when all he was trying to do was demonstrate that an earlier comment was unsubstantiated.
    The point, Kim, is that any conclusion about relative MCAS performance drawn from the data Jeff posted is invalid. The data cannot be used to either prove or disprove the Town Crier poster's assertion. The data doesn't suggest the opposite to be true, the data suggests nothing whatsoever.

    Why would all such conclusions be invalid? Because there are any number of uncontrolled variables besides experiencing or not experiencing reconfiguration that could have influenced the MCAS results produced by those groups of students. Either Jeff didn't understand this rather basic point when he posted, in which case Probability and Statistics 101 is in order, or he understood it but posted this data to induce readers to reach an invalid conclusion that supported his position.

  3. #48
    Join Date
    Nov 2005
    Posts
    726

    Default

    Quote Originally Posted by Dave Bernstein View Post
    The point, Kim, is that any conclusion about relative MCAS performance drawn from the data Jeff posted is invalid. The data cannot be used to either prove or disprove the Town Crier poster's assertion. The data doesn't suggest the opposite to be true, the data suggests nothing whatsoever.

    Why would all such conclusions be invalid? Because there are any number of uncontrolled variables besides experiencing or not experiencing reconfiguration that could have influenced the MCAS results produced by those groups of students. Either Jeff didn't understand this rather basic point when he posted, in which case Probability and Statistics 101 is in order, or he understood it but posted this data to induce readers to reach an invalid conclusion that supported his position.
    Dave, my head hurts. Some guy said something on the Crier that I believe we are all agreeing was unproven. Jeff tried to point that out [i.e., the point of his post was that the assertion could not be proven] and now you're telling him he can't prove or disprove the assertion. I took a lot of statistics at a pretty mathematical college, and my occupation is in applying causal logic to understanding system behavior. And with this background, Dave, I'm still completely baffled by your logic.

  4. #49
    Join Date
    Jan 2007
    Location
    Wayland, MA
    Posts
    235

    Default

    Quote Originally Posted by Kim Reichelt View Post
    Dave, my head hurts. Some guy said something on the Crier that I believe we are all agreeing was unproven. Jeff tried to point that out [i.e., the point of his post was that the assertion could not be proven] and now you're telling him he can't prove or disprove the assertion. I took a lot of statistics at a pretty mathematical college, and my occupation is in applying causal logic to understanding system behavior. And with this background, Dave, I'm still completely baffled by your logic.
    I say that Jeff can't disprove the Crier poster's assertion with the data he presented. Please review this data in

    http://www.waylandenews.com/forum/sh...ol-MCAS-scores

    There were four groups of students, two of which endured reconfiguration, two of which did not.

    Hypothesis: the only factor responsible for the differences in MCAS results achieved by these groups is whether a group did or did not endure reconfiguration.

    Do you agree or disagree with this hypothesis?

    In his post, Jeff pointed out that groups from one school outperformed groups from another school. Is this not evidence of at least one other results-affecting factor in play?

    Suppose you were tasked with designing an experiment that would reliably (say, 90% confidence) predict the impact of reconfiguration on MCAS results achieved by elementary students. What would this experiment look like? How would you eliminate the affects of differences in teacher performance, differences in home life, differences in activities before/during the tests, etc. in order to achieve the desired confidence level?

    Now compare Jeff's experiment to the one you designed. How would you characterize the confidence level of his results?

  5. #50
    Join Date
    Nov 2005
    Posts
    726

    Default

    Quote Originally Posted by Dave Bernstein View Post
    I say that Jeff can't disprove the Crier poster's assertion with the data he presented. Please review this data in

    http://www.waylandenews.com/forum/sh...ol-MCAS-scores

    There were four groups of students, two of which endured reconfiguration, two of which did not.

    Hypothesis: the only factor responsible for the differences in MCAS results achieved by these groups is whether a group did or did not endure reconfiguration.

    Do you agree or disagree with this hypothesis?

    In his post, Jeff pointed out that groups from one school outperformed groups from another school. Is this not evidence of at least one other results-affecting factor in play?

    Suppose you were tasked with designing an experiment that would reliably (say, 90% confidence) predict the impact of reconfiguration on MCAS results achieved by elementary students. What would this experiment look like? How would you eliminate the affects of differences in teacher performance, differences in home life, differences in activities before/during the tests, etc. in order to achieve the desired confidence level?

    Now compare Jeff's experiment to the one you designed. How would you characterize the confidence level of his results?
    Dave, if Jeff wasn't trying to prove the poster's assertion wrong, but was merely trying to assert that the poster's assertion was unproven, would you be OK with that, and could we call it a night?

  6. #51
    Join Date
    Jan 2007
    Location
    Wayland, MA
    Posts
    235

    Default

    Quote Originally Posted by Kim Reichelt View Post
    Dave, if Jeff wasn't trying to prove the poster's assertion wrong, but was merely trying to assert that the poster's assertion was unproven, would you be OK with that, and could we call it a night?
    1. Jeff ended the Elementary School MCAS Scores thread by acknowledging that the data he posted did not rule out a drop in scores among those students enduring reconfiguration, which means that the data does not "suggest otherwise" with respect to the Town Crier poster's assertion as he originally claimed. Anyone can confirm this by reading messages 91 and 92 of that thread; why he disavows it now is fodder for those still undecided as to his character.

    2. Having claimed knowledge of statistics and logic, you've said here that you're completely baffled by my logic. I've responded with the exposition above. If there are errors in the case I've laid out, please describe them; otherwise, please have the courtesy to acknowledge its correctness.

  7. #52
    Join Date
    Nov 2005
    Location
    Wayland MA
    Posts
    1,431

    Default

    Quote Originally Posted by Dave Bernstein View Post
    1. Jeff ended the Elementary School MCAS Scores thread by acknowledging that the data he posted did not rule out a drop in scores among those students enduring reconfiguration, which means that the data does not "suggest otherwise" with respect to the Town Crier poster's assertion as he originally claimed.
    The illogic of this statement stuns me.

    A. MCAS scores improved coincident with the elementary school reconfiguration (which I described using the word suggest), but because those scores involved different cohorts and were affected by factors other than the reconfiguration, it's not possible to prove that the reconfiguration caused the score increase.

    B. As I agreed, the reconfiguration might have had a detrimental affect on scores countered by some other stronger positive force.

    Statements A and B on my part aren't contradictory. As I've stated all along, I'm not aware of any data that either counters or proves the suggestion.

    Quote Originally Posted by Dave Bernstein View Post
    Anyone can confirm this by reading messages 91 and 92 of that thread; why he disavows it now is fodder for those still undecided as to his character.
    I'm disavowing nothing. Why you continue to raise my character (analytical, thoughtful, consistent, responsive) as an issue is curious.

  8. #53
    Join Date
    Nov 2005
    Posts
    726

    Default

    Jeff beat me to the punch on replying. Dave, I am not arguing that Jeff's analysis is a proof of anything regarding the impact of the reconfiguration on MCAS scores, merely that the post on the Town Crier discussion boards that instigated the whole discussion was itself not supported by available information.

    Seriously, why are you turning down every available opportunity to find common ground?

  9. #54
    Join Date
    Jan 2007
    Location
    Wayland, MA
    Posts
    235

    Default

    Quote Originally Posted by Jeff Dieffenbach View Post

    A. MCAS scores improved coincident with the elementary school reconfiguration (which I described using the word suggest), but because those scores involved different cohorts and were affected by factors other than the reconfiguration, it's not possible to prove that the reconfiguration caused the score increase.

    B. As I agreed, the reconfiguration might have had a detrimental affect on scores countered by some other stronger positive force.

    Statements A and B on my part aren't contradictory.
    Jeff, your original assertion was

    Quote Originally Posted by Jeff Dieffenbach View Post
    With respect to the Town Crier poster's allegations that MCAS scores slipped and that the reconfiguration was the cause, the data suggests otherwise.
    Given your acknowledgement in B above, your original assertion can only remain true if the phrase suggests otherwise is redefined to mean can induce an incorrect conclusion.

    You have again agreed that the data you posted does not refute the Town Crier poster's claim. So why did you post that data and suggest that it does?

    Quote Originally Posted by Jeff Dieffenbach View Post
    As I've stated all along, I'm not aware of any data that either counters or proves the suggestion.
    Give us a break, Jeff; it took 92 posts for you to acknowledge that the data you posted doesn't disprove the Town Crier poster's claim. Had you appended "I'm not aware of any data that either counters or proves the suggestion" to your your first post, the only question would have been "so what's the point of your posting data that you say proves nothing either way?"

    Quote Originally Posted by Jeff Dieffenbach View Post
    Why you continue to raise my character (analytical, thoughtful, consistent, responsive) as an issue is curious.
    Either you did not realize at the time that "some other stronger positive force" or forces might have been in play, or you did and proceeded anyway. The latter possibility is what raises the character issue.
    Last edited by Dave Bernstein; 05-12-2010 at 03:15 PM.

  10. #55
    Join Date
    Jan 2007
    Location
    Wayland, MA
    Posts
    235

    Default

    Quote Originally Posted by Kim Reichelt View Post
    Jeff beat me to the punch on replying.
    Yes, as a tag team, you two should improve your footwork.

    Quote Originally Posted by Kim Reichelt View Post
    Dave, I am not arguing that Jeff's analysis is a proof of anything regarding the impact of the reconfiguration on MCAS scores, merely that the post on the Town Crier discussion boards that instigated the whole discussion was itself not supported by available information.

    Seriously, why are you turning down every available opportunity to find common ground?
    I thought that we had reached agreement when Jeff finally acknowledged (after 92 posts) that the data he posted did not refute the Town Crier poster's claim. Having made that effort, you wouldn't expect me to simply ignore his recent disavowal, would you?

    You have called my logic into question, but failed to either point out its flaws or acknowledge its correctness. I'm certain that had you found an error of any significance, you'd have reported it by now. Do you not understand what is now expected in civil discourse?

  11. #56
    Join Date
    Nov 2005
    Posts
    726

    Default

    Quote Originally Posted by Dave Bernstein View Post
    You have called my logic into question, but failed to either point out its flaws or acknowledge its correctness. I'm certain that had you found an error of any significance, you'd have reported it by now. Do you not understand what is now expected in civil discourse?
    The flaw is that as I understood it you were telling Jeff he hadn't proven something that, as I understand it, he wasn't attempting to prove. He was merely attempting to demonstrate that another something hadn't been demonstrated.

    As for my not understanding what is expected in civil discourse, were you being funny? (If so, sorry, end here with a smiley :-) if not, then...) I had thought I did understand -- that what is expected in "civil discourse" is that people have discourse (interaction, conversation, communication) which is civil (that is polite and demonstrating common courtesy). (I did, in fact, delete some draft comments I had almost posted in this thread involving the picking of nits which I thought were witty, but might have been misconstrued as being uncivil). Perhaps you have some other set of criteria in mind?

    But I feel like I'm repeating myself here, and not adding anything, so I'll step off of this one unless there's something you guys need me for.
    Last edited by Kim Reichelt; 05-12-2010 at 04:04 PM.

  12. #57
    Join Date
    Jan 2007
    Location
    Wayland, MA
    Posts
    235

    Default

    Quote Originally Posted by Kim Reichelt View Post
    The flaw is that as I understood it you were telling Jeff he hadn't proven something that, as I understand it, he wasn't attempting to prove. He was merely attempting to demonstrate that another something hadn't been demonstrated.
    Jeff attempted to refute the Town Crier Poster's assertion by presenting data that at the time he claimed "suggests otherwise". In point of fact, the data he posted neither supports nor contradicts the poster's assertion: as Jeff has acknowledged once again, other factors or forces besides reconfiguration might have influenced the test results.

    Do you you agree with this summary?

    Had Jeff's initial post been along the lines of "yes there is MCAS data, but there are so many factors in play besides reconfiguration that no conclusions can be drawn", there'd have been no debate. Such a response would have been far weaker than the data-driven approach Jeff tried to take, but since Jeff's data conveys no actual information with respect to reconfiguration impact, it was equally weak once it's lack of information was exposed.

    Quote Originally Posted by Kim Reichelt View Post
    As for my not understanding what is expected in civil discourse, were you being funny?
    No. When you challenge someone's logic or reasoning in a non-specific way - e.g. " I'm still completely baffled by your logic" - and they respond with an explanation, it's expected that you will either point out the flaws, or retract your challenge.

  13. #58
    Join Date
    Nov 2005
    Location
    Wayland MA
    Posts
    1,431

    Default

    I'll try again.

    1. Anonymous said that ES reconfiguration caused an MCAS DROP
    2. I could NOT find data that showed an MCAS drop
    3. I COULD find data that SUGGESTED an MCAS increase
    4. Because of other factors (for instance, different cohorts), I could not PROVE an MCAS increase
    5. As a result, I could not rule out an MCAS drop caused by ES reconfiguration or other factors

    In sum, the data is more suggestive of improvement than decline, but does not prove anything either way. From my perspective, it's rare to find real-world examples were causation can be proved. That does not mean, however, that we should stop looking for evidence. Having some idea is better than having no idea at all, especially if we keep in mind that we don't have the luxury of perfect information.

  14. #59
    Join Date
    Nov 2005
    Posts
    726

    Default

    Quote Originally Posted by Dave Bernstein View Post
    Had Jeff's initial post been along the lines of "yes there is MCAS data, but there are so many factors in play besides reconfiguration that no conclusions can be drawn", there'd have been no debate.
    I thought that was what he did... For example, Jeff's first three quotes from that other thread (bold emphasis below is mine)

    Quote Originally Posted by Jeff Dieffenbach View Post
    The original poster's other fallacy--confusing correlation with causation--also applies here. I'm not suggesting that the reconfiguration was the cause of improved MCAS performance. Rather, I place the credit squarely on the shoulders of our excellent administrators and teachers who elevated their game under difficult circumstances.
    and

    Quote Originally Posted by Jeff Dieffenbach View Post
    Actually, I did not provide a theory for the ES reconfiguration, I provided an observation: namely, that MCAS scores improved more in the reconfiguration year than in the year prior.
    and

    Quote Originally Posted by Jeff Dieffenbach View Post
    That's not a theory *about* MCAS scores and reconfiguration, that's an "anti-hypothesis" suggesting that I don't find any evidence to link reconfiguration to MCAS scores.
    Finally, to respond to your other point:

    Quote Originally Posted by Dave Bernstein View Post
    When you challenge someone's logic or reasoning in a non-specific way - e.g. " I'm still completely baffled by your logic" - and they respond with an explanation, it's expected that you will either point out the flaws, or retract your challenge.
    I thought I explained multiple times my understanding of events, which had Jeff not claiming to prove something but seeking to demonstrate another comment unproven. That's still where I'm at, so if you're trying to prove some other point, it's starting to feel like we're speaking two different languages. Maybe the disconnect involves the use of the word "suggest", which I take in this context to mean "works in the same direction as, but doesn't prove", and maybe you interpret some other way?

    Maybe the solution to bridging our gap is defining some English words, rather than going back over Jeff's posts again and again. Is "suggest" the problematic word, or might there be some other word? Or can I just stop contributing to this thread?

  15. #60
    Join Date
    Jan 2007
    Location
    Wayland, MA
    Posts
    235

    Default

    Quote Originally Posted by Jeff Dieffenbach View Post
    I'll try again.

    1. Anonymous said that ES reconfiguration caused an MCAS DROP
    2. I could NOT find data that showed an MCAS drop
    3. I COULD find data that SUGGESTED an MCAS increase
    4. Because of other factors (for instance, different cohorts), I could not PROVE an MCAS increase
    5. As a result, I could not rule out an MCAS drop caused by ES reconfiguration or other factors

    In sum, the data is more suggestive of improvement than decline, but does not prove anything either way. From my perspective, it's rare to find real-world examples were causation can be proved. That does not mean, however, that we should stop looking for evidence. Having some idea is better than having no idea at all, especially if we keep in mind that we don't have the luxury of perfect information.
    Yes, getting solid evidence can be difficult. that's not a valid excuse for suggesting conclusions based on shoddy evidence.

    Had you included the above sentence "In sum, the data is more suggestive of improvement than decline, but does not prove anything either way" (along with a brief summary of the many uncontrolled factors responsible for this uncertainty) in your original post, there would have been no debate. Instead, you played fast and loose with the data, and the word "suggests".

    73,

    Dave, AA6YQ

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •