Unexpected Serving Size



  •  Okay, we've all seen stupid stuff like cans of pop or candy bars that claim they contain two servings - just so they can claim a lower calorie count per serving.  However, we spotted this one at dinner last night, and it seemed like an odd candidate for such a trick, unless they're concerned about the sodium content... This is on a jar of "Great Gherkins" spear dill pickles.




  • Serving sizes used for nutrition info reporting are regulated by the FDA, precisely to stop shenanigans like 8 oz servings in a 12 oz can. Pickles are not one of the categories where the serving size is defined as the nearest number of pieces to the reference amount. Your picture fits perfectly the "1 oz (28 g/visual unit of measure)" rule given in the table - they couldn't say "1 spear (1.3 oz/38g)" if they wanted to.



  • They really should be required to publish the info for the entire container, and then if they want they can also add the "per serving" nonsense as another column in the table. Who actually drinks 250ml of a 483ml can, and who can be arsed to figure out the proper numbers from that?



  •  @lolwtf said:

    They really should be required to publish the info for the entire container, and then if they want they can also add the "per serving" nonsense as another column in the table. Who actually drinks 250ml of a 483ml can, and who can be arsed to figure out the proper numbers from that?

     

    There's plenty of cases where giving stats on the whole container is pretty meaningless. For example, who eats a jar of peanut butter at a time? What they need to do is tighten down the regulation of what a "serving" is in absurd cases.



  • @Maciej said:

    For example, who eats a jar of peanut butter at a time?

    He does!



  • @lolwtf said:

    They really should be required to publish the info for the entire container, and then if they want they can also add the "per serving" nonsense as another column in the table. Who actually drinks 250ml of a 483ml can, and who can be arsed to figure out the proper numbers from that?
     

    I think here in the UK that actually is the rule. Looking at a Pepsi can I happen to have here, the table has two columns, one labelled "per 100ml" and one labelled "per 330ml" (can size).

    BTW below this is another table showing "Guideline Daily Amounts" of "Calories" (2000), "Fat" (70g), "Saturates" (20g), "Sugar" (90g) and "Salt" (6g).



  • If you know what "serving size" really is, this always makes perfect sense.

    It's really "sample size". You send some amount of your food to the lab. They divide it into N equally-sized samples, test them for content, and average the results. That's your nutrition information.

    So when you have a 483 ml can with a serving size of 250 ml, the reason is quite probably that the test lab uses 250 ml beakers for liquid samples.

    The purpose of nutrition information isn't really for you to know what nutrition you get when you eat it, anyway, but for you to compare similar kinds of products and see which one is more to your liking. So it's not important whether the sample size represents an actual serving, it's much more important that the sample size represent the same across multiple products.



  • @lolwtf said:

    They really should be required to publish the info for the entire container, and then if they want they can also add the "per serving" nonsense as another column in the table.

    There's no need to reinvent the wheel like that. The FDA could simply copy or adapt the system used in the UK (and I think across the EU), where food which needs a nutritional label always has quantities of protein, carbohydrates, fats and energy in quantity per 100g to the nearest calorie or tenth of a gramme. This is really useful on a 11g pack of nori (seaweed).



  •  Maybe there not very good pickles?  Perhaps people take one bite (potentially 3/4 of it) and decide they dont want pickles anymore?



  • @MadnessASAP said:

     Maybe there not very good pickles?  Perhaps people take one bite (potentially 3/4 of it) and decide they dont want pickles anymore?

    Who would eat 3/4 of one of these things?  Who *could* eat 3/4 of one of these things at one go?

    Sweet Gherkins are really the way to go.  Bonus: 1 pickle is 1 serving, at 25-28oz, depending on the brand.  And there's much less sodium.



  • @_moz said:

    There's no need to reinvent the wheel like that. The FDA could simply copy or adapt the system used in the UK (and I think across the EU), where food which needs a nutritional label always has quantities of protein, carbohydrates, fats and energy in quantity per 100g to the nearest calorie or tenth of a gramme.
     

    It's like that in Australia too. Nutritional information listed per 100mL or 100g, as well as "per serving". I haven't noticed a non-integer number of seving per packet/can/etc for a while though.

    I remember a ages ago that Weight Watchers margarine advertised "30% less fat per serving". But their serving size was 3.5g and every other margarine serving was 5g. Of course using 30% less will reduce fat by 30%!

     



  • Fat free cooking spray!* 0 grams of fat per serving!†

    (* Sole ingredient: canola oil)

    († Serving size: 1/3 of a second spray)



  • <FONT color=#c0c0c0>@TwelveBaud said:

    1/3 of a second spray
    Is this trying to take account of the fact that the first spray (after being unused after a non-zero amount of time) results in nothing coming out?</FONT>



  • @CDarklock said:

    If you know what "serving size" really is, this always makes perfect sense.

    It's really "sample size". You send some amount of your food to the lab. They divide it into N equally-sized samples, test them for content, and average the results. That's your nutrition information.

    So when you have a 483 ml can with a serving size of 250 ml, the reason is quite probably that the test lab uses 250 ml beakers for liquid samples.

    And exactly why would they need to report the information only for the sample size used in the lab?  It may be that it's more difficult to measure a large sample than a small one, so dividing it up into small samples makes sense.  Fair enough.  However, if the average of 100 samples of 250 ml each indicates that one sample contains 137 calories, it's simple math to arrive at the result of a 483 ml can containing 265 calories.  So it makes no sense to report random sizes.

    @CDarklock said:

    The purpose of nutrition information isn't really for you to know what nutrition you get when you eat it, anyway, but for you to compare similar kinds of products and see which one is more to your liking. So it's not important whether the sample size represents an actual serving, it's much more important that the sample size represent the same across multiple products.
    This is true, but (at least here in Finland) 100 ml is far more common for that.  It's also convenient as a base for scaling, as it happens to be a power of ten. 



  • No, I don't expect that the labels should ever show the lab's sample size. Different tests for different substances (fat, vitamins etc) will probably have different sample sizes and that would really stuff theings up. (Not to mention different labs having different procedures.)

    The serving sizes are most appropriate to larger packets. Does a can of soft drink sold in a pack of 6 contain more or less kilojoules than an average glass of orange juice from a 2-litre bottle? Does the one that claims it's got lots of vitamin C actually give you more vitamin C than the other one? In most metric countries, I believe the standard cup measure is 250ml.

    Per 100g or 100ml makes it easy to compare on a percentage basis. (Remember, in the metric system, 100g = 100ml for pure water and the conversion is close enough for most edible substances.) This allows you to compare products regardless of the manufacturer's serving size or pack size. If there's 1 gram of fat in 100ml, then you can say that the product is 1% fat.

    You will always get certain inconsistencies, like the rounding error that allows the cooking oil to be rounded off to zero or pickle servings not quite matching the size of the pickles that are in the jar. So long as there is enough information for the consumers to make their puchasing decisions without a pocket calculator, the system is probably good enough.



  • @tdb said:

    And exactly why would they need to report the information only for the sample size used in the lab?

    Need? No, no, no.

    They report the sample size used in the lab, because that is what the lab's report contains. They simply duplicate the lab's report. They do this because they are required by law to print nutrition information.

    They don't actually care whether you get useful information. The usefulness of the information is largely an accident caused by the lab's desire to use similar size samples across products, which simplifies their operations.



  • As I said above, they are also [i]required by law[/i] to report the nutrition information for - e.g. for pickles specifically - a 1 oz (28g) serving and to state approximately how much [i.e. 3/4 spear] 1 oz (28g) is. They don't get to decide to use a bigger serving size just because you feel like it. All this talk about lab sample sizes is complete nonsense. The serving sizes used for these labels are tightly regulated by the FDA, and the fact that you don't find the choices they made to be "aesthetic" enough doesn't change that.

    Oh, and guess what? The king-size candy bars and large (16-oz or half-liter) cans [b]do[/b] contain two servings, no matter what you tell yourself to justify eating it all in one go, fatass.



  •  Perhaps we can read one of the links you referenced:

    "The serving size of products that come in discrete units... is the number of whole units
    that most closely approximates the reference amount."

    So it would seem to me, that 1 pickle is 1 serving size.

    What is interesting is they give this example "(28g/about 1/2
    pickle)"

     

    Looking at the table:

    Fruits and Fruit Juices:
        Candied or pickled11.......  30 g............  -- piece(s) (-- g)
     
    Ohhh it is actually a Veggie, so we have:
        Pickles, all types11.......  30 g............  1 oz (28 g/visual
                                                        unit of measure)
     
    WTF? I can say 1 piece (38g) if it was a pickled fruit, but I have to say 1 oz (3/4 of a piece) because it is a pickled veggie?
     

     



  • @chrismcb said:

    "The serving size of products that come in discrete units... is the number of whole units
    that most closely approximates the reference amount."

    So it would seem to me, that 1 pickle is 1 serving size.

    Yes, if you assume that pickles "come in discrete units". They don't. They come in jars.



  • @chrismcb said:

        Candied or pickled11.......  30 g............  -- piece(s) (-- g)

    The application failed to initialize properly (0xc0000135): pickle.dll



Log in to reply
 

Looks like your connection to What the Daily WTF? was lost, please wait while we try to reconnect.