SailNet Community - View Single Post - Duncan Loop in lieu of eye splice or bowline?
View Single Post
post #15 of Old 04-06-2011
Senior Slacker
SlowButSteady's Avatar
Join Date: Feb 2010
Location: SoCal
Posts: 4,223
Thanks: 0
Thanked 17 Times in 15 Posts
Rep Power: 8
Originally Posted by AdamLein View Post
What was wrong with their representation? I mean, if you're going to complain about something, why not the tiny sample size?
When I teach BioStatistics I spend about two and a half lectures (almost an entire week of class) on data presentation, so I could give you a pretty long list of the errors made in the presentation cited.

But, to start (since you ask):

- "Figures" 1, 2, and 3 are tables, not figures
- Standard Errors (in addition to Standard Deviations) should be given, as we are interested in a comparison of means
- Units should be SI, not American/English customary (however, engineers never seem to understand this)
- The Abstract is a disaster (don't get me started)
- The Equipment and Methods sections (essentially analogous to a "Materials and Methods" section of a standard format scientific paper) should be expanded with greater detail
- Figures (and Tables) are presented but not cited in the text
- Figure 8 is presented as if the different knots have some ordinal relationship, which they do not.
- Figures are inconsistent as to orientation (strength is scaled horizontally in Figs. 4 - 7 and vertically in Fig. 8)
- Fig. 7 is redundant, it simply repeats means presented in Figs. 4 - 6
- All of the figures present means, yet none have error bars (SD or SE)
- Differences in knot breaking strength, as it relates to material strength, are discussed (albeit briefly) without any tests of statistical significance
- And (one of my pet peeves), the word "data" is the plural of "datum". Therefore, "the data was" is grammatically somewhat akin to saying, "the cows was," (and sounds like fingernails on a chalkboard to Yours Truly)

As to the small sample size; it's actually probably OK for such a study. The variation about the mean here appear small enough that increasing the replication would probably not change the over-all results enough to worry about.

Never forget them. Do something to prevent it from happening again.
Charlotte Bacon, Daniel Barden, Rachel Davino, Olivia Josephine Gay, Ana M. Marquez-Greene, Dylan Hockley, Dawn Hochsprung, Madeleine F. Hsu, Catherine V. Hubbard, Chase Kowalski, Jesse Lewis, James Mattioli , Grace McDonnell, Anne Marie Murphy, Emilie Parker, Jack Pinto, Noah Pozner, Caroline Previdi, Jessica Rekos, Avielle Richman, Lauren Rousseau, Mary Sherlach, Victoria Soto, Benjamin Wheeler, Allison N. Wyatt

Last edited by SlowButSteady; 01-02-2012 at 08:29 PM.
SlowButSteady is offline  
Quote Share with Facebook
For the best viewing experience please update your browser to Google Chrome