A basket of 24 staples.
In my earlier article I discussed the BBC's reporting of some research done by mysupermarket.co.uk on food prices. I omitted detailed discussion of the research and the validity of the conclusions based upon it because it wasn't relevant there, however some people may be interested so I include that here.
We base this on 24 items and portion sizes enough for a family of four for a weekly shop (items listed below and full breakdown attached). … Almost all of the items were own branded where possible, but to make it a fair comparison we had to chose products that were most alike in all three supermarkets, and available during those specific months.
- Personal email from mysupermarket.co.uk
From this description I deduce that they selected their baskets after the period in question. My scientific principles tell me that it would be much better if they had selected them before the period started as then their experiment cannot be biased by the data. However I have no evidence that they have in fact, consciously or unconsciously, selected things in order to support any particular hypothesis.
The contents of their baskets are slightly strange however:
Now I have two problems with this:- firstly how was that list selected as a representative list of staple foodstuffs purchased weekly, in particular I can't even imagine getting through 500g of Sugar in a week; and secondly, even assuming that that is a representative list of staples, how valid is it to generalise and assume that the prices in it are representative for an entire £100 weekly shop as they do in the section of the email quoted in the other post.
We base this on 24 items and portion sizes enough for a family of four for a weekly shop (items listed below and full breakdown attached). … Almost all of the items were own branded where possible, but to make it a fair comparison we had to chose products that were most alike in all three supermarkets, and available during those specific months.
- Personal email from mysupermarket.co.uk
From this description I deduce that they selected their baskets after the period in question. My scientific principles tell me that it would be much better if they had selected them before the period started as then their experiment cannot be biased by the data. However I have no evidence that they have in fact, consciously or unconsciously, selected things in order to support any particular hypothesis.
The contents of their baskets are slightly strange however:
- 800g White Loaf
- Packet of 7 or 8 "Kids" Bananas
- Bag of Golden Delicious Apples
- 3 Mixed Peppers
- A Cucumber Portion
- An Iceburg Lettuce
- 6 Tomatoes
- 2.5kg Maris Piper Potatoes
- 6pts Semi-Skimmed Fresh Milk (single bottle, I think)
- 12 Medium Free Range Eggs
- 250g Salted/Creamy Butter
- 500g Fresh British Beef Mince
- 170g Bernard Matthews Wafer Thin Turkey Ham
- 900g-1kg Garden Peas
- 420g Baked Beans
- 500g Dolmio Original Bolognese Sauce
- 454g Strawberry Jam
- 500g Silver Spoon Half Spoon Granulated Sugar
- 500g Corn Flakes
- 500g Fusilli Pasta
- 1kg Basmati Rice
- 80 Tea Bags
- 1l Orange Juice
Now I have two problems with this:- firstly how was that list selected as a representative list of staple foodstuffs purchased weekly, in particular I can't even imagine getting through 500g of Sugar in a week; and secondly, even assuming that that is a representative list of staples, how valid is it to generalise and assume that the prices in it are representative for an entire £100 weekly shop as they do in the section of the email quoted in the other post.