Food waste is a global issue, accounting for 8% of the world’s greenhouse gas emissions. In comparison, aviation accounts for just over 2%. If food waste were a country, it would be the third biggest emitter after the US and China. According to the FAO, we waste one third of all food produced for human consumption. But this statistic is from 2011. It could be even worse now, or it could be better. But how do we measure food waste to begin with?
Today, 29 September 2020, is the UN’s International Day of Awareness of Food Waste and Food Loss. However, at present there is no standard way of measuring food waste across countries, within academia, or even across different organisations in the same country. How are we meant to be aware of the extent of food waste and food loss if the method of measuring it is inconsistent, or worse, inaccurate?
In the UK, the largest amount of food waste occurs at the household level. I am most interested in this area, as it means the household level is where the most potential lies to reduce food waste. Methods to measure quantity of food waste from households vary from questionnaires and interviews, to food waste diaries and even physical waste surveys, where food waste is extracted from disposal units and weighed. They each have advantages and disadvantages, and crucially, some can be biased in ways that even risk compromising the study.
Surveys and questionnaires
The biggest benefit of using surveys and interviews is the number of participants you can involve. As an easy and efficient research method, large sample sizes can be administered, sometimes gathering thousands of responses. This is the method commonly used by WRAP (Waste and Resources Action Plan) in the UK.
However, surveys and questionnaires also have the biggest drawbacks. Significantly, they are self-reporting research methods, meaning there is a high risk of participant bias. Especially when it comes to food waste, participants are likely to underreport how much food they actually throw away. One element of this is what Van Herpen et al. (2019) call “social desirability”. It is a well-known phenomenon in social science that participants tend to want to be viewed in a positive light, even by researchers who are strangers.
In addition, throwing food away is such a common and mindless activity, participants are likely to underreport or overreport the quantity simply because they don’t realise or remember how much they are discarding, even if they try and be truthful (Hebrok and Boks, 2017). I wouldn’t be able to tell you how much food I throw away in grams, and exactly what that food was – it is difficult to remember and to quantify what you’ve discarded in retrospect. In one recent study, most participants believed that they do not waste food at all, or only a little (Giordano et al., 2018), and in another, participants underreported their food waste by a tenfold (Delley and Brunner, 2018).
| Risk of bias | Sample size | Usefulness* | Burden for participant | Cost/burden for researcher |
| High | Large | Medium | Low | Low |
*Usefulness for measuring the quantity of food waste.
Interviews
Interviewing takes more time and energy than surveys and questionnaires, so the sample size is likely to be smaller. However, especially with semi-structured interviews where questions can be adapted for each participant, more detail can be extrapolated and participants can be asked to explain their answers more fully.
Nevertheless, this method is also a type of self-reporting, and biases are common. Participants will shape their answers based on what they think the interviewer wants to hear, or position themselves in a way they believe is more favourable. This disadvantage may decrease if multiple interviews are conducted over time and particularly if interviews are done in the participants’ homes as “ethnographic interviews” (see Evans 2014), but this adds to the invasiveness and burden for participants.
Interviews are more useful for answering questions about why people waste food rather than the quantity of food wasted, and are usually deployed alongside other methods. To assess data, interviews will have to be transcribed and coded, which can increase burden for the researcher.
| Risk of bias | Sample size | Usefulness | Burden for participant | Cost/burden for researcher |
| Medium | Medium | Low | Medium | Medium |
Food waste diaries
A common tool for food waste studies, participants record their food waste over days or weeks in a diary. This can include measuring their waste in grams or describing the food in detail. Some ask participants to give further details, like whether the food was leftovers, how mouldy it had become, what time of day it was thrown away etc. This has a great benefit for measuring the quantity of food waste at household level and reasons behind it, and better still, measures these over a longer time period that surveys and questionnaires.
Nevertheless, as well as being vulnerable to underreporting bias, keeping a diary can influence genuine changes to behaviour over the time period of the study. For example, the diary may make participants less wasteful if they want to improve their reporting or realise how much they are throwing away.
Diaries are also a lot of effort for the participant and have high drop-out rates, even over just one week (Langley et al., 2010). Therefore, even if a large sample size was achieved at the beginning, this may reduce considerably by the end of the study.
| Risk of bias | Sample size | Usefulness | Burden for participant | Cost/burden for researcher |
| Medium | Medium | High | High | Medium |
Photo elicitation
Photo elicitation uses photos to prompt discussion with a research participant, although almost no food waste studies have used this method (see Farr-Wharton et al., 2014).
For my masters dissertation, I combined photo elicitation with in-depth interviews with the participants in their homes. This meant taking a photo of their fridge after they did a grocery shop, while they talked about their food habits, and taking another a week later to compare. Using these photos, I conducted long interviews about what happened to the food that week and why.
The benefits of using photos is that the evidence is there right in front of the research participant and it is harder for biases to influence the collection of data. A photo is an ‘unmediated and unbiased visual report’ (Schwartz 1989:120).
This technique is used in the hospitality industry. For example, a tech company in London called Winnow is using “smart bins” to measure food waste in restaurants. The bins take a photo of food whenever it is dropped into it and identifies what type of food it is.
However, of course participants could choose not to use the smart bin if they wanted to influence the data, and one of my own participants admitted that they purchased certain foods because they knew I was coming to take a photo of their fridge that day.
Moreover, the time needed to code and analyse photos can be significant for the researcher. One study went further and looked at whether photos could be coded by participants themselves to discern the volume of food waste. However they concluded that due to the wide variation of overestimation and underestimation, participants would benefit from training before reporting their waste (van Herpen and van der Lans, 2019). Either that, or a comprehensive database would be useful to help participants measure up their waste against some examples. In both cases, burdens would increase for participant and researcher.
| Risk of bias | Sample size | Usefulness | Burden for participant | Cost/burden for researcher |
| Medium | Small | High | Medium | Medium |
Physical waste survey
In a physical waste survey, the researcher measures the quantity of food waste in the food bin, dumpster or whatever disposal unit the participant uses. Although a messy business, the researcher can sort the waste into categories tailored to their study (e.g. type of food like bread or vegetables, type of waste like leftovers or whole foods). This is Rathje and Murphy’s (2001) preferred method for this reason.
There is no room for participant bias in this method, as all waste is recorded by a third party, unless the participant engages in radical behaviour like refusing to throw their food waste down the toilet instead.
The main disadvantage is this method is heavily dependent on resources. It is time and energy intensive, requires multiple trained researchers, and sample sizes will be small unless more resources are afforded.
Other practical problems arise. Elimelech et al. (2018) point out that it is harder to differentiate between individual households if food waste in multi-apartment buildings where food waste is collected together in a shared bin. They add that as organic matter, food degrades quickly, and if not examined in time, may not be discernible or amalgamate with other foods. Food waste fed to pets or put in compost will also be omitted, unless diaries or interviews are conducted alongside.
| Risk of bias | Sample size | Usefulness | Burden for participant | Cost/burden for researcher |
| Low | Medium | High | Low | High |
Which method is best?
The following table summarises the different methods.
From this, it transpires that surveys/questionnaires and physical waste surveys are the best methods for measuring the quantity of food waste. Interviews, food waste diaries and photo elicitation may be more useful for exploring why food is wasted or other qualitative questions.
Surveys and questionnaires can garner greater sample sizes, are low burden for participants, and can be resource efficient for researchers. This is reflected in how government-funded organisations like WRAP use them most often. However, there is great risk of under or over-reporting and participant bias.
On the other hand, physical waste surveys have little risk of bias, are extremely useful for assessing the quantity of food waste and are little burden to the participant. The main drawback is the amount of time and resource needed to complete them, so their success is entirely depended on the institution’s or researcher’s capacity.
Elimelech et al. (2019) helpfully conducted a study of 192 households in which they used both questionnaires and physical waste surveys. Shockingly, they found that only 6% of participants gave an accurate report of their food waste in the questionnaires, compared to what was calculated in physical waste surveys.
It is therefore highly recommended to pair surveys or questionnaires with another method of data collection to verify the answers, or strengthen the quality of data, as they are not valiable methods of data collection on their own for measuring quantity of food waste.
Can food waste be globally standardised?
Withanage et al. (2020) analysed food waste studies between 2010 and 2019 and weighed up more advantages and disadvantages (omitting photo elicitation). They concluded that the ‘notable issue with having such diverse array of methods is that the researchers are unable to compare the food waste scenarios for two different geographical regions at a higher accuracy’ (p. 13).
Standardisation is needed across borders when it comes to quantifying food waste, not just in terms of how it is measured, but the classifications of what types of food, and what condition they are in at the point of disposal.
Different nations will have different food cultures, diets and disposal habits, and some will have wildly different levels of food waste. In theory, standardisation will help uncover which countries are similar in their wastage, help tailor prevention strategies, and compare which strategies work well. Standardisation will also be critical for measuring progress. But will it work in practice? Nobody has ever tried, and this may be a key question for the FAO.
Another fundamental question must be answered, namely: what is food waste to begin with? Some people consider potato skins as very much part of the food, and some consider them inedible parts that have to be discarded. Are potato skins therefore food waste, or not? These perceptions will differ within countries as well as between them. Who decides what food waste is, and what are the political, economic and environmental consequences of the definition? I explore these questions in another long read.
References
- Delley, M. and Brunner, T. A., 2018. Household food waste quantification: comparison of two methods. British Food Journal. 120(7). pp. 1504-1515.
- Elimelech, E., Ayalon, O. and Ert, E., 2018. What gets measured gets managed: A new method of measuring household food waste. Waste Management, 76. pp. 68-81.
- Elimelech, E., Ert, E. and Ayalon, O., 2019. Bridging the gap between self-assessments and measured household food waste: A hybrid valuation approach. Waste Management, 95. pp. 259-270.
- Evans, D., 2014. Food Waste: Home Consumption, Material Culture and Everyday Life. London: Bloomsbury.
- Farr-Wharton, G., Foth, M. and Hee-Jeong Choi, J., 2014. Identifying factors that promote consumer behaviours causing expired domestic food waste. Journal of Consumer Behaviour, 13. pp. 393-402.
- Giordano, C., Piras, S., Boschini, M. and Falasconi, L., 2018. Are questionnaires a reliable method to measure food waste? A pilot study on Italian households. British Food Journal, 120(12). pp. 2885-2897.
- Langley, J., Yoxall, A., Heppell, G., Maria Rodriguez, E., Bradbury, S., Lewis, R., Luxmoore, J., Hodzic, A. and Rowson, J., 2010. Food for Thought? – A UK pilot study testing a methodology for compositional domestic food waste analysis. Waste Management and Research, 28(3). pp. 220-227.
- Rathje, W. and Murphy, C., 2001. Rubbish! The Archaeology of Garbage. New York: HarperCollins.
- Schwartz, D., 1989. Visual Ethnography: Using Photography in Qualitative Research. Qualitative Sociology, 12(2). pp. 119-154.
- Van Herpen, E. and van der Lans, I., 2019. A picture says it all? The validity of photograph coding to assess household food waste. Food Quality and Preference, 75. pp. 71-77.
- Van Herpen, E., van der Lans, I. A., Holthuysen, N., Nijenhuis-de Vries, M., Quested, T. E., 2019. Comparing wasted apples and oranges: An assessment of methods to measure household food waste. Waste Management, 88. pp. 71-84
- Withanage, S. V., Maria Dias, G. and Habib, K., 2020. Review of household food waste quantification methods: Focus on composition analysis. Journal of Cleaner Production, 279. pp. 1-15.
FOOD WASTE STORIES 