Are Male Students Served More Food Than Female Students in EJ?

ejorigin
6 days ago
6 min read

Written by: Justa (24-E4), Julianna (24-E6),Kai Rui (24-I1), Jia Xuan (24-I6), Marco (24-O1), Skyler (24-U2), Joshua (24-U4), Kristen (24-U6)

Same dish. Same price. Same portions? Our investigation might prove otherwise.

The Eunoian canteen follows the standard procedure for any school canteen. During peak lunch hours, canteen vendors can expect to serve students continuously for hours, with little to no breaks. See the student, smile, serve the student, collect payment, rinse and repeat.

Is that all there is to it? That’s what we believed, over here at Press, until we began to observe an interesting phenomenon. It seemed that the canteen vendors would take into account other factors when serving us food. On a few occasions, when our female friends would purchase food, our male friends would appear with the same dish, only with more (e.g. double the serving of fried chicken). Was this mere coincidence? Were we imagining things?

We thus embarked on an investigation to determine whether male students were served more food than female students in school.

The preparation and ingredients: How we did it

For this experiment, we weighed the briyani set meal from the Malay stall, and the chicken katsu don from the Japanese stall. As such, any conclusion drawn only applies to these two dishes.

Over a few days, we asked (begged) people to weigh their food using an electronic weighing scale we left by the microwaves. We then collated the data (food mass, gender, dish, date, time) using a Google form, and used Excel to analyse it.

We surveyed people by reaching out via word of mouth and towards the end, by randomly approaching everyone we saw who bought the food. Overall, we received 21 female and 16 male respondents for biryani and 12 female and 11 male respondents for the chicken katsu don.

H₀: Mean mass of food given to males = Mean mass of food given to females

H_a: Mean mass of food given to females < Mean mass of food given to males

In math terms, the null hypothesis (or status quo), H₀, was that the mean mass of food given to males equal to that given to females, given that the dish and the price was constant. Our alternative hypothesis, H_a, was that the mean mass of food given to females is less than the mean mass of food given to males. Our level of significance was decided at 5%.

We used Excel to conduct a two-sample t-test assuming unequal variance on the data. A t-test computes the t-statistic — the difference between the means of the two samples subtracting the differences between the population means (0 in our case since the status quo is that both means are the same), divided by the standard error. The standard error measures how much you expect the sample means to vary just by random chance.

(xA is the sample mean for females, xB is the sample mean for males; 𝝁A is the population mean for females, 𝝁B is the population mean for males; sA is the standard deviation for females, sB is the standard deviation of males; nA is number of females, nB is the number of males)

This shows how big the difference the sample means was compared to the expected random variation. Afterwards, we find the probability (p-value) of getting a difference at least this large just by chance. A small p-value means it’s unlikely the difference is just random.

The kitchen skills: The assumptions

Data values are independent. Measurements for one observation do not affect measurements for any other observation
Data in each group must be obtained via a random sample from the population (not the case here)
Data in each group is normally distributed (questionable in our study)
Data values are continuous

The final dish: Our results

Chicken katsu don

Our p value is 0.285 which is significantly higher than our level of significance which is 5%. Thus we do not reject the null hypothesis and cannot conclude that males get more food than females.

Briyani

Our p-value is 0.000304 which is less than our level of significance which is 5%. Thus, we reject H_o and conclude that males do get more food than females.

This is what the weights for briyani look like in a Box and Whisker plot (blue is female and red is male):

The food review: The limitations

Again, the sample is NOT random! Rather, the sample might just be the reflection of our social ties and who we managed to convince to do this experiment. Most of the respondents were J2s, with some repeated people in our dataset. Also other factors like the length of the queue and the time of day (peak vs off-peak, morning vs afternoon when they’re trying to get rid of the food left) also influence the amount of food you get. Some people in our survey could also be extremely loyal to a particular store and, if they become friends with the vendor, may get more food than average.

For the Japanese store specifically, people were served in 2 different bowls of different masses: one black and one red. While we did subtract the mass of the bowls away, we only measured the masses of the bowls once and could have measured an outlier. In addition, the sizes of the bowls are slightly different which could affect the amount of food served (studies have shown that people eat more when the plate is bigger so a similar effect for bowls is not far fetched). Lastly, the sample size for the Japanese store is small, which means we cannot definitely say that it is an accurate reflection of all food weights from there.

Though an imperfect experiment, the data for briyani still highlight a statistically significant difference in the food weights for the same dish and the results should not be dismissed out of hand.

The chef’s story: Reasons why

We began with the hypothesis that food amount is determined by gender, which can be a valid reason as to the difference in food — the average male requires more daily calories than the average female, so it would make sense to serve them more food. Another possibility we considered was that food portions are ascertained by size, wherein larger portions are served to those who have larger stature.

Furthermore, this could also be done because of purely economic reasons, where store owners may ration their food to serve more students (and earn more profits!). To them, those who are smaller sized will naturally eat less and perhaps even throw the food away if they are served too much.

The discernment between who gets more food leaves the question of whether it was the food server’s choice to make. If all students are paying a fixed price, should they all receive the same amount of food or can the choice to eat less be made by the server?

On a larger scale, this may reflect the “growing boy” standards of most societies. As they age through puberty, boys may be told to “eat more, grow stronger”, while girls are not expected to make significant changes to their diets. If anything, diet culture influences girls to eat less. Ultimately, female students getting served less food could mirror a larger, systemic issue wherein patriarchal forces want girls to be thinner and smaller. The first line of attack: their food intake.

Food coma: Conclusion

Whether you see portion sizes between the genders as mere coincidence or not, it is important to note that we do not wish this to serve as an attack on our canteen stall vendors. It takes a lot of hard work and dedication to cater to our 1500+ EJ community, with their long hours beginning early in the morning. Our beloved canteen vendors do not have any nefarious intentions, and should not be treated as such.

Instead, we hope this article provides you with a new perspective on the food you’re served. There are intrinsic biases that go into the decisions that allow you to taste every delicious morsel of your meal, and it is important that we begin to unpack the why.

And for our female readers, maybe consider letting your guy friends order your food next time you’re extra hungry.

Compliments to the kitchen: Acknowledgements

Kristen’s friend for help on data analytics and suggesting the two sample t-test when what we learnt was the z-test.

Everyone who weighed their food! We are eternally grateful for letting us influence your food choices for a meal or two for our experiment.

For further reading

On the t-test: