The FastDay Forum

[Moogie]

Just a couple of little updates this evening, but that's two more things crossed off my list!

- Your progress charts page will now display your starting BMI and how many BMI points you've lost.

- The forum stats page now includes the standard deviation (can someone mathematical please tell me if the figure looks right, I've just a pre-built function and applied it to the array of losses so I think it should be right, but no idea what sort of figure it should be giving!). So far this figure is only visible when you filter the stats, but the main forum stats will include it once they have automatically regenerated (Doctor Who style!) tonight.

Tonight I intend to continue work on the long awaited and much delayed recipes area - but the best laid plans, and all that... we'll see

You rock Moogie x

AWESOME!!! Thanks so much.

Looks fantastic, thanks!

Your amazing thankyou.

Oh I love that - really encouraging - thanks so much for adding it

Hi Moogie,

I'm an analytical chemist which sort of counts as mathematical I suppose.

Your standard deviations (SD) number looks within the realm of possibility. It means that the data points vary from the average by 0.335kg (don't quite remember the number).

You may want to post the relative standard deviation (RSD) which is expressed as a percentage. It's just the SD / AVG so for your data the RSD is about 28% which means the data points vary by about 30%.

This can be very encouraging for someone since that is a pretty high RSD and illustrates what we all know: results from person to person will vary! Some will lose fast, some slow but the average remains the same.

I would be interested in seeing the RSD of the average values over time....meaning how much variance is there from the average 1lb per week?

Hope this helps.

Thank you Moogie for your hard work. My waist measurement chart had looked as though it was having a nervous breakdown but has now recovered! All the other information is very useful although the number of weeks until I reach my goal weight caused a depressed gasp!

HI Moogie good work

I spent 25 years doing materials research; while not a statistician I spent about a quater of my time using stats to analyse my data. So I think I had a fair idea of stats; but that was over ten year ago !!!

A few points come to mind:

The average and standard deviation should be in the same units; for me the average is in lbs and SD in Kgs.

The standard devation is a measure of variation in the data. It seems in the right order; average is 1.48 lbs SD 0.74 lbs
The data is made up from different groups; men, women, 4:3, 5:2 ,ect so will not form a normal distribution (bell shaped). We have to be very careful in using the SD for any simple tests or statements as most assume a normal distribution.

Adding the Mode as well as the Medium would be useful; the mode gives the most quoted data; ie the weight loss most people have per month.

I hope this is helpful and my memory as held up

That is brilliant thank you

[Moogie]

Thanks jack, it should be displaying in the same units, so I guess I balled up on a variable name somewhere. Will bear your other suggestions in mind too, ta. Remember the SD will also display when you filter the results so you can limit the scope of the data.

Looks just great Moggie, You are doing a fantastic job here - THANKS so much

jacksati wrote "medium" but he meant "Median". Stephen Jay Gould gave many lectures about using the median rather than the mean (or "average"), whenever the data must always be positive or above some level. People can never be below zero kg, so zero kg forms an absolute boundary to this data. This is why we should use the median.

The median is the 50th percentile. Half the people will be above it, and half will be below it.

You can still say "the average person", but you should always use the median to calculate that average.

I am so impressed! Nice job, Moogie!

[Moogie]

jacksati wrote "medium" but he meant "Median". Stephen Jay Gould gave many lectures about using the median rather than the mean (or "average"), whenever the data must always be positive or above some level. People can never be below zero kg, so zero kg forms an absolute boundary to this data. This is why we should use the median.

The median is the 50th percentile. Half the people will be above it, and half will be below it.

You can still say "the average person", but you should always use the median to calculate that average.

We already have the medians displayed on the tracker