Matt Kuklis

08-03-2004, 05:26 AM

BIOMCH-L folks,

Thanks so much for those of you who provided input with respect to my

question regarding scaling population percentiles. Most of the

respondents had some disclaimer related to not directly being able to

accurately scale population data as well as if one does scale he/she must

understand the results as to not misuse them. It is certainly my

understanding that when dealing with ergonomic data, it is always best to

err on the side of conservative. Thanks again for the feedback!

Please see below for the original question and a summary of responses.

Matt Kuklis

Biomechanical Engineer

Hill-Rom Co., Inc.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Original Email (Question):

Group,

A colleague and I are working on a project to try to understand a

comfortable range of push/pull forces that a 5th percentile female can

endure. We have found many tables of forces published by many individuals

mostly to do with maximum endurance forces for various push/pull

exercises. We found a decent representation of our application in

Dreyfuss' "Measure of Man and Woman". However, the forces given are for a

50th percentile male. Dreyfuss gives a male to female scaling factor of

2/3; we've also seen scaling factors around 1/2. We are interested in the

5th percentile female as I have mentioned.

Is there any such scaling factor/conversion that can take data (in

particular push/pull forces) from one segment of the population (based on

percentile) to another (e.g., is there a conversion to take data from 50th

percentile to 5th percentile)?

Short of doing our own study, is there a better way to get the data more

directly? For instance, is there a more gold-type standard/study of

push/pull forces for various exercises (i.e., forearm/shoulder in

different orientations; sitting versus standing, etc.)?

A summary of responses will be posted. Please indicate if you would like

your name removed from the summary posting.

Thank you for any information.

Matt Kuklis

Biomechanical Engineer

Hill-Rom Co., Inc

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Hi Matt:

I am not sure what data you have access to, but if the Dreyfuss data are

true percentile data then they can be represented by a normal distribution curve. The 50th percentile will correspond to a Z-score of 0.0. Using this distribution,

you can find the Z-score that corresponds to the 5th percentile female by looking up a conversion in a Standard Normal Probability table (see for example, http://www.math2.org/math/stat/distributions/z-dist.htm, http://psych.colorado.edu/~mcclella/java/normal/normz.html, http://davidmlane.com/hyperstat/z_table.html).

Using this conversion, you will see that the 5th percentile on a normal curve corresponds to a Z-score of about -1.6452. If you have the standard deviation of the measures used by

Dreyfuss, you can then calculate the actual mean values for a 5th percentile female by solving:

Z = (Y-MEAN)/SD

where Z = the z-score (-1.6452); Y = the 5th percentile female value (to be solved); MEAN = the population mean (the Dreyfuss female 50th percentile/mean force); and SD is the female

population standard deviation.

I hope that works. I don't know another way to do it. Let me know how it comes out.

Peter

______________________________________

Peter F. Vint, Ph.D.

Research Scientist

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Matt,

You need to look at the Snook Liberty Mutual Psychophysical Tables.

Carter Kerk

Carter J. Kerk, PhD, PE, CSP, CPE

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Hi Matt

Muscle forces are generally proportional to muscle cross-sectional area,

so to approximately (body mass)^2/3. You could use that to convert data

from male subjects to females, if you knew the body mass of both groups.

However, although that would account for the effects of size, it

wouldn't allow for any sex differences.

I don't think you can convert from the 50th percentile to the 5th

percentile without knowing what distribution the data follow and having

a measure of their variability.

Hope that's some help. Good luck with your work,

Sharon

------

Sharon R. Bullimore

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

These articles do evaluate push/ pull.

Much of Snook's work uses psychophysical information. The articles

probably will

not give you conversions that allow adjusting data from the 50th

percentile male to accommodate the 5th percentile female. The 1990

article in ergonomics does consider females as well as males.

Ciriello VM. Snook SH. Hughes GJ. Further studies of psychophysically

determined maximum acceptable weights and forces. [Journal Article] Human

Factors. 35(1):175-86, 1993 Mar.

Snook SH. Ciriello VM. The design of manual handling tasks: revised

tables of maximum acceptable weights and forces. [Journal Article]

Ergonomics. 34(9):1197-213, 1991 Sep.

Ciriello VM. Snook SH. Blick AC. Wilkinson PL. The effects of task

duration on psychophysically-determined maximum acceptable weights and

forces. [Journal Article] Ergonomics. 33(2):187-200, 1990 Feb.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

There are no valid scaling methods available to transform a 50th

percentile

person to a 5th percentile person. The reason for this is that the concept

of a "5th percentile person" is neither biologically nor statistically

valid.

Thomas M. Greiner, Ph.D.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Hello Matt,

I would agree with the previous speaker about converting data. Better to

obtain a better data source. I have here on my bookshelves a book called

"Adultdata", published by the Dept of Trade and Industry in the UK. It has

a

whole raft of tables related to different pushing/pulling situations, and

gives 5% and 50% values for males and females.

I think you can get a free copy if you send them a mail.

Good luck,

Jan Herman Kuiper

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Matt

You talk about "enduring" forces. Do you meant that the force is being

applied to the person and they have tor resist it or do you mean that they

have to exert that force for a set period of time?

I'm surprised you don't seem to have found Snook's psychophysical data

which includes maximum acceptable pushing and pulling forces for males and

females. He quotes values down to 10th percentile so you will need to

calculate the SD from his figures to calculate 5th percentile.

Also, I've not seen a copy of Dreyfuss lately but that data is now very

old, and probably originally from the US military so will be biased

towards

young fit males

The suggested female:male strength ratio of 2:3 is almost meaningless.

Steve Pheasant pointed out in 1983 that there is actually tremendous

variability in it.

You might also find the U Michigan "3D Static Strength Prediction Program"

of use - it can predict the percentage of the population capable of

exerting a given force in a given direction in a given posture. Sorry,

but

I don't have the URL to hand but a search of the umich.edu website should

find it/

Some suggested references:

Snook SH and Ciriello V (1991) "The design of manual handling tasks:

Revised tables of maximum acceptable weights and forces". Ergonomics 34,

1197-1213

Pheasant ST (1983) "Sex differences in strength - some observations on

their variability" Applied Ergonomics 14, 205-211

A recent report by a couple of my colleagues: RR228 - Review of the risks

associated with pushing and pulling heavy loads

http://www.hse.gov.uk/research/rrhtm/rr228.htm

Regards

Andrew

Dr Andrew Pinder, PhD, MSc, Eur Erg, MErgS

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Hi Matt,

The Health and Safety Executive in the UK has just recently done a review

of the risks associated with pushing and pulling. Quite a bit of

information on pushing / pulling forces is brought together, and you find

it helpful in directing you to other sources. It is Research Report

RR228,

"Review of risks associated with pushing and pulling heavy loads", which

can be downloaded free from the internet (

http://www.hse.gov.uk/research/rrhtm/201-300.htm ) or purchased from 'HSE

Books'.

Alternatively, the Department of Trade and Industry in the UK have

produced

two free books called "Strength Data for Design and Safety": Phase 1 and

Phase 2. (obtained through www.dti.gov.uk I believe or by calling +44 20

7215 0383). It presents data according to age for a range of pushing and

pulling tests, which may be of help.

Good luck,

Jeremy Ferreira

-----------------------------------------------------------------

To unsubscribe send SIGNOFF BIOMCH-L to LISTSERV@nic.surfnet.nl

For information and archives: http://isb.ri.ccf.org/biomch-l

Please consider posting your message to the Biomch-L Web-based

Discussion Forum: http://movement-analysis.com/biomch_l

-----------------------------------------------------------------

Thanks so much for those of you who provided input with respect to my

question regarding scaling population percentiles. Most of the

respondents had some disclaimer related to not directly being able to

accurately scale population data as well as if one does scale he/she must

understand the results as to not misuse them. It is certainly my

understanding that when dealing with ergonomic data, it is always best to

err on the side of conservative. Thanks again for the feedback!

Please see below for the original question and a summary of responses.

Matt Kuklis

Biomechanical Engineer

Hill-Rom Co., Inc.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Original Email (Question):

Group,

A colleague and I are working on a project to try to understand a

comfortable range of push/pull forces that a 5th percentile female can

endure. We have found many tables of forces published by many individuals

mostly to do with maximum endurance forces for various push/pull

exercises. We found a decent representation of our application in

Dreyfuss' "Measure of Man and Woman". However, the forces given are for a

50th percentile male. Dreyfuss gives a male to female scaling factor of

2/3; we've also seen scaling factors around 1/2. We are interested in the

5th percentile female as I have mentioned.

Is there any such scaling factor/conversion that can take data (in

particular push/pull forces) from one segment of the population (based on

percentile) to another (e.g., is there a conversion to take data from 50th

percentile to 5th percentile)?

Short of doing our own study, is there a better way to get the data more

directly? For instance, is there a more gold-type standard/study of

push/pull forces for various exercises (i.e., forearm/shoulder in

different orientations; sitting versus standing, etc.)?

A summary of responses will be posted. Please indicate if you would like

your name removed from the summary posting.

Thank you for any information.

Matt Kuklis

Biomechanical Engineer

Hill-Rom Co., Inc

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Hi Matt:

I am not sure what data you have access to, but if the Dreyfuss data are

true percentile data then they can be represented by a normal distribution curve. The 50th percentile will correspond to a Z-score of 0.0. Using this distribution,

you can find the Z-score that corresponds to the 5th percentile female by looking up a conversion in a Standard Normal Probability table (see for example, http://www.math2.org/math/stat/distributions/z-dist.htm, http://psych.colorado.edu/~mcclella/java/normal/normz.html, http://davidmlane.com/hyperstat/z_table.html).

Using this conversion, you will see that the 5th percentile on a normal curve corresponds to a Z-score of about -1.6452. If you have the standard deviation of the measures used by

Dreyfuss, you can then calculate the actual mean values for a 5th percentile female by solving:

Z = (Y-MEAN)/SD

where Z = the z-score (-1.6452); Y = the 5th percentile female value (to be solved); MEAN = the population mean (the Dreyfuss female 50th percentile/mean force); and SD is the female

population standard deviation.

I hope that works. I don't know another way to do it. Let me know how it comes out.

Peter

______________________________________

Peter F. Vint, Ph.D.

Research Scientist

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Matt,

You need to look at the Snook Liberty Mutual Psychophysical Tables.

Carter Kerk

Carter J. Kerk, PhD, PE, CSP, CPE

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Hi Matt

Muscle forces are generally proportional to muscle cross-sectional area,

so to approximately (body mass)^2/3. You could use that to convert data

from male subjects to females, if you knew the body mass of both groups.

However, although that would account for the effects of size, it

wouldn't allow for any sex differences.

I don't think you can convert from the 50th percentile to the 5th

percentile without knowing what distribution the data follow and having

a measure of their variability.

Hope that's some help. Good luck with your work,

Sharon

------

Sharon R. Bullimore

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

These articles do evaluate push/ pull.

Much of Snook's work uses psychophysical information. The articles

probably will

not give you conversions that allow adjusting data from the 50th

percentile male to accommodate the 5th percentile female. The 1990

article in ergonomics does consider females as well as males.

Ciriello VM. Snook SH. Hughes GJ. Further studies of psychophysically

determined maximum acceptable weights and forces. [Journal Article] Human

Factors. 35(1):175-86, 1993 Mar.

Snook SH. Ciriello VM. The design of manual handling tasks: revised

tables of maximum acceptable weights and forces. [Journal Article]

Ergonomics. 34(9):1197-213, 1991 Sep.

Ciriello VM. Snook SH. Blick AC. Wilkinson PL. The effects of task

duration on psychophysically-determined maximum acceptable weights and

forces. [Journal Article] Ergonomics. 33(2):187-200, 1990 Feb.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

There are no valid scaling methods available to transform a 50th

percentile

person to a 5th percentile person. The reason for this is that the concept

of a "5th percentile person" is neither biologically nor statistically

valid.

Thomas M. Greiner, Ph.D.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Hello Matt,

I would agree with the previous speaker about converting data. Better to

obtain a better data source. I have here on my bookshelves a book called

"Adultdata", published by the Dept of Trade and Industry in the UK. It has

a

whole raft of tables related to different pushing/pulling situations, and

gives 5% and 50% values for males and females.

I think you can get a free copy if you send them a mail.

Good luck,

Jan Herman Kuiper

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Matt

You talk about "enduring" forces. Do you meant that the force is being

applied to the person and they have tor resist it or do you mean that they

have to exert that force for a set period of time?

I'm surprised you don't seem to have found Snook's psychophysical data

which includes maximum acceptable pushing and pulling forces for males and

females. He quotes values down to 10th percentile so you will need to

calculate the SD from his figures to calculate 5th percentile.

Also, I've not seen a copy of Dreyfuss lately but that data is now very

old, and probably originally from the US military so will be biased

towards

young fit males

The suggested female:male strength ratio of 2:3 is almost meaningless.

Steve Pheasant pointed out in 1983 that there is actually tremendous

variability in it.

You might also find the U Michigan "3D Static Strength Prediction Program"

of use - it can predict the percentage of the population capable of

exerting a given force in a given direction in a given posture. Sorry,

but

I don't have the URL to hand but a search of the umich.edu website should

find it/

Some suggested references:

Snook SH and Ciriello V (1991) "The design of manual handling tasks:

Revised tables of maximum acceptable weights and forces". Ergonomics 34,

1197-1213

Pheasant ST (1983) "Sex differences in strength - some observations on

their variability" Applied Ergonomics 14, 205-211

A recent report by a couple of my colleagues: RR228 - Review of the risks

associated with pushing and pulling heavy loads

http://www.hse.gov.uk/research/rrhtm/rr228.htm

Regards

Andrew

Dr Andrew Pinder, PhD, MSc, Eur Erg, MErgS

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Hi Matt,

The Health and Safety Executive in the UK has just recently done a review

of the risks associated with pushing and pulling. Quite a bit of

information on pushing / pulling forces is brought together, and you find

it helpful in directing you to other sources. It is Research Report

RR228,

"Review of risks associated with pushing and pulling heavy loads", which

can be downloaded free from the internet (

http://www.hse.gov.uk/research/rrhtm/201-300.htm ) or purchased from 'HSE

Books'.

Alternatively, the Department of Trade and Industry in the UK have

produced

two free books called "Strength Data for Design and Safety": Phase 1 and

Phase 2. (obtained through www.dti.gov.uk I believe or by calling +44 20

7215 0383). It presents data according to age for a range of pushing and

pulling tests, which may be of help.

Good luck,

Jeremy Ferreira

-----------------------------------------------------------------

To unsubscribe send SIGNOFF BIOMCH-L to LISTSERV@nic.surfnet.nl

For information and archives: http://isb.ri.ccf.org/biomch-l

Please consider posting your message to the Biomch-L Web-based

Discussion Forum: http://movement-analysis.com/biomch_l

-----------------------------------------------------------------