Navigation – Plan du site

AccueilNuméros57-58Hors dossierUsing Q-methodology to Understand...

Hors dossier

Using Q-methodology to Understand The Human Interactions with Pollinating Insects in Mayotte (French Island, Indian Ocean of East Africa)

Utilisation de la méthodologie Q pour comprendre les interactions humaines avec les insectes pollinisateurs à Mayotte (île française, océan Indien d'Afrique de l'Est)
Samuel Perichon

Résumés

En 2021, le gouvernement français a lancé un nouveau plan national pour lutter contre le déclin des populations d'insectes pollinisateurs. Sa déclinaison à Mayotte (Océan Indien) représente une réelle opportunité d'améliorer les connaissances scientifiques sur ces êtres vivants et sur les liens que l'homme entretient avec eux. Cela d’autant plus que, localement, très peu de recherches ont été conduites sur ce sujet. Notre étude repose sur l’application de la Méthodologie Q. Cette méthodologie permet d’étudier les points de vue subjectifs des individus à partir d’un tri préférentiel – dans notre cas, il s’agissait de 39 photos en lien avec les insectes pollinisateurs – dans une distribution de type Gaussien. Les tris effectués (n = 29) ont permis d’identifier quatre profils intégrant à des degrés divers des considérations écologiques, économiques et culturelles dans la perception des insectes et de leurs services écosystémiques. A l'échelle de l'archipel, le profil où les considérations écologiques sont prégnantes ne concerne sans doute qu’une part minoritaire de la population, car il a été principalement développé par des individus ayant fait des études supérieures. Dans les autres profils, les insectes pollinisateurs sont ignorés, voire redoutés ou détestés. L'abeille domestique est une espèce à part, non seulement parce qu'elle fournit aux hommes le miel, un produit naturel dont les propriétés médicinales sont reconnues localement, mais aussi parce qu'un verset du Coran lui est consacré. Avec 95% de la population se déclarant musulmane, la religion contribue donc probablement à la valorisation de l'abeille.

Haut de page

Notes de l’auteur

Acknowledgements

The author would like to thank all those who agreed to take part in our survey. Thanks to Lisa, Fissou, Mouhamadi, Youssrah et Ismaël for their support during the Q-sorts and discussions. To Pierre Baby, for his advice on entomology. To the members of the board of the Mahoran Federation of Environmental Associations (FMAE). To Garrett Dempsey for his proofreading of the article.

Texte intégral

Introduction

1With over a million described species, entomofauna accounts for 80% of the world's known animal species (Jactel et al., 2020). Of the twenty-eight orders in this class of invertebrates, four - Coleoptera, Diptera, Hymenoptera and Lepidoptera - account for more than half of all insect species (Chapman, 2009). These four orders include the main pollinators, the most emblematic of which are bees, bumblebees and butterflies. Their role is essential to the functioning of (agro-)ecosystems. Almost 90% of flowering plants depend on them, and 84% if we consider the three hundred plants grown in Europe (Allsopp et al., 2008). In terms of global agricultural production tonnage, this represents around 35% of what we consume each year (Klein et al., 2007). Across Europe, bee and butterfly populations have fallen by 37% and 31% respectively in just a few years, and one in ten of these two groups of pollinators is now threatened with extinction (Pott et al., 2016). The causes of insect decline are well documented, and are many and often cumulative: climate change, pollution, urbanisation, invasive alien species, intensive agriculture, deforestation, drought, fire, etc. (Wagner et al., 2021). Aware of the situation, public authorities in some countries have already regulated practices and customs in favour of entomofauna, and are encouraging initiatives to promote and raise awareness among the general public, particularly of the ecosystem services provided by pollinating insects. In France, a national action plan has been renewed in 2021 for a period of five years, with the specific aim of halting this decline and improving scientific knowledge (French government, 2021).

2Mayotte's implementation of the National Plan for Insect Pollinators is a real opportunity, not least because of a lack of knowledge about local interactions between humans and insects. That integrates traditional environmental knowledge, which is passed on orally but can be threatened by social change. Among other things, it is based on a collection of social representations related to pollinating insects. The fact that there has been no study on this subject to date reinforces the importance of collecting information from the population of Mayotte. We chose a methodology that has never been used in this region, the Q methodology (Stephenson, 1935), to accomplish this. It is based on ranking items according to agreement or preference in the case of photographs, and explaining their ranking during a follow-up interview (Dieteren et al., 2023). This article details the results of photo sorting and factorial analysis, which revealed four sorting profiles. The study site, material, and methods selected will be presented.

1. Site of study: the Mayotte archipelago

3Mayotte is a small volcanic archipelago in the Indian Ocean, 300 km north-west of Madagascar and 450 km from the coast of Mozambique (Figure 1). The small archipelago covers an area of 374 km² and comprises two main islands: Grande-Terre (363 km²) and Petite-Terre, as well as islets such as Mtamboro, Mbouzy, Brandrélé and the islet of Sable Blanc. Mayotte is surrounded by a 1,500 km² lagoon, making it one of the largest on the planet. The lagoon itself is surrounded by a 160 km long coral reef cut by a dozen passes. The archipelago is thought to have been formed nine million years ago on the landward side, with an estimated genesis of fifteen million years ago.

Figure 1. The study site

Figure 1. The study site

4The topography of Mayotte is characterised by slopes. Two-thirds of the surface area of Grande-Terre is marked by slopes of over 15%, at altitudes above 300 metres (Oberlinkels et al., 2007). Four mountain ranges structure this area, including the Bénara massif, whose eponymous peak is the highest point on the island (653 m). These massifs are subject to the monsoon winds (north to north-west), which explain the significant variations in rainfall between the highlands in the north and the coastline in the south-east. The maritime tropical climate is characterised by a fairly narrow annual and daily temperature range, and abundant rainfall (>1,500 mm) concentrated between November and April. Forests and agroforestry systems cover 10,792 ha, the equivalent of 30% of the island's surface area, but only 8% of this surface area corresponds to forests little disturbed by human activities1. Mayotte remains one of the tropical islands with the greatest species richness in relation to its surface area. In its "statistics on the biogeographical status of recorded species", the French National Inventory of Natural Heritage (INPN) lists: 6,061 native species, 265 endemic species, 272 introduced species, including 40 invasive species, and 321 domestic species2. Again according to the INPN, more than 900 species have protected status, but 435 are currently threatened with extinction. The French Committee of the International Union for Conservation of Nature believes that this threat could affect 43% of native plants3. Mayotte is on the list of thirty-four global biodiversity hotspots.

5As far as pollinating insects are concerned, it is difficult to state with precision, how many species are present. Based on the data available on the INPN website, we can nevertheless venture an estimate. It is likely that 300 to 350 species could be pollinators. Very few of them provide three of the four ecosystem services: "regulation", "culture" and "supply". The honey bee is one of these. It's worth noting that Mayotte has both traditional gathering practices and modern beekeeping for honey supply. Our study showed that beekeeping for a long time consisted of harvesting one or more honeycombs from cavity trees in agricultural plots and doing what they can to keep the bees there. Local associations and public establishments have recently started offering training courses in conventional beekeeping. The number of beekeepers using divisible hives and marketing their honey production is estimated to be over a hundred. This change in beekeeping practices is accompanied by a different type of packaging for local honey (jar with label). Consumers who are used to coarsely filtered honey sold in small plastic bottles or plastic bags are often upset by this. They comprehend that this jarred honey does not match traditional harvesting practices, resulting in its lower value in medicinal properties (Perichon, 2023). Futhermore, the Muslim religion encourages this attachment in the minds of believers, with a verse in the Koran dedicated to the benefits of bees for mankind.

6Mayotte's population is still largely Afro-Malagasy and animist, with Islam playing a significant role in its culture (Dauphin & Thibault, 2011). Shimaoré, a Bantu language that is related to Swahili, is the most widely spoken language. Shibushi, a variation of the Malagasy language, is also spoken in some communities. Young people, graduates, and people from metropolitan France are the main speakers of French, the official language. The region is under increasing pressure from human activity, due to the sustained population growth fuelled by migration and a high fertility rate. Mayotte had a population of more than 250,000 in 2017, which was a four-fold increase from 67,200 in 1985. On 1 January 2022, the population is estimated to be 300,000. The population could rise by a factor of 2.5 by 2050, as per projections. The level of education is still low, with two-thirds of the working population having few or no qualifications, compared to 20% nationally. In the second quarter of 2022, the unemployment rate stood at 34 percent. These high rates of unemployment can partly be explained by the underdeveloped business sector. In Mayotte, living standards have deteriorated over the past ten years, with 77% of the population now below the poverty line (national level) and 42% living on less than €160 a month. Crime is on the rise, causing a feeling of insecurity. At least one burglary or theft without breaking and entering occurred in 18% of households in Mayotte in 2018 or 2019; theft from agricultural plots (vegetables, fruit, animals) was claimed by 30% of respondents. Agriculture is still a significant activity for a third of households (with a 19% share of the total surface area of the archipelago). Traditional combined cropping systems are used grown in small areas (0.45 ha on average) with a tree density of more than 50 trees per hectare on half the cultivated land. These traditional systems, known as "mundra" (analogous to "Creole gardens"), offer a number of benefits, including: preventing erosion on sloping land that is often exposed to this risk, ensuring production spreads throughout the year, promoting food self-sufficiency and dietary diversity, etc.

2. Material and methods

7The Q Methodology is still widely used in social science research as one of the research methods used to analyze individuals' subjective points of view (Dieteren et al., 2023; Rost, 2020; Watt & Stenner, 2012; Brown, 1980). Sorting items into a forced distribution is what Q-sort is defined as. The items in our case were photos. All the sorted items have been analyzed through a web application which we will describe shortly. By taking this step, we were able to identify profiles of viewpoints without necessarily having to link them to socio-professional or socio-demographic variables.

2.1. Steps in the Q Methodology

8We followed the Q Methodology's five steps in our approach (Table 1). The Q sample, or the sample of photos to be sorted, was one of the most crucial steps. To find relevant photos, specialists, particularly in entomology, and research were necessary, given that thirty-nine were to be selected for the survey. We favoured a standard range from -4 for the least popular photos to +4 for the most popular, resulting in a distribution in nine columns (-4, -3, -2, -1, 0, +1, +2, +3, +4). The interest of the Q-sort also lies in the fact that the preferences are part of a Gaussian-type distribution (Figure 2). Thirty-nine photos were selected for distribution. Five themes related to pollinating insects were depicted in the photos: l) "Pollinating insects and their environment" are represented by fourteen photos, twelve of which show an insect in close-up, most often foraging. Lepidoptera has the largest number of taxa in the images, with seven photos. The environment of the pollinating insects shows a natural habitat, a dense forest of Mayotte seen from the sky, and a semi-natural habitat representative of the island's agricultural plots; 2) "Anthropogenic threats to pollinating insects" appear four times in the Q sample, three of which concern agriculture (deforestation, use of fire and pesticides); 3) There are ten photos on "Honey gathering and modern beekeeping"; 4) "Honey, wax and pollen" covers five photos ; and 5) "The dangers of Hymenoptera for humans" are essentially linked to the consequences of insect stings.

Table 1: List of photographs and summary sorting by factor

Synthesis sorting

Consens us rank

Q sample

I

II

III

IV

1

20

A blue butterfly (Junonia rhadama: female) on a branch

1

1

-2

2

2

6

A Vinson's butterfly (Papilio demodocus) foraging on a bougainvillea

2

1

0

2

3

2

A black and red butterfly (Acraea ranavalona: male) on a leaf

1

1

1

2

4

18

A Little Monarch (Danaus chrysippus) foraging on a purple flower

3

0

-1

1

5

23

A Striped Policeman (Coeliades forestan) foraging an orange flower

2

-1

-2

1

6

21

A Heteroceros (Achaea sp.) on a wall

-1

-1

-4

0

7

7

A heterocere (Lymantriinae sp.) on a branch

-1

-2

-1

-3

8

22

A mango flower beetle (Protaetia fusca) on white flowers

1

-2

-3

-2

9

30

A wasp with an orange abdomen (Sphex torridus) on a field flower

2

-1

-4

-1

10

31

A mason wasp (Delta emarginatum) on a purple flower

2

-1

0

-4

11

27

A bee (Apis mellifera) foraging on a yellow flower (dandelion)

4

2

-1

1

12

28

A carpenter bee (Xylocopa caffra: female) foraging on a mauve flower

1

2

-3

0

13

8

A colony of bees having built a dozen combs under a branch

3

2

4

4

14

9

A colony of bees having built a dozen combs inside a dwelling (under a roof)

0

-2

0

-2

15

14

A cluster of bees under a branch near a house

1

0

-2

-2

16

11

A young man walking down the street with a swarm of honey bees on his arm

-2

-2

0

-3

17

19

An allergic reaction in the face after a Hymenoptera sting

-2

-4

-3

0

18

5

A bee sting with a stinger stuck in the forearm

-1

0

-2

-1

19

29

About fifteen combs spread out on the ground. Some of the combs are smoking, burnt and bleached by the smoke.

-3

0

1

-4

20

24

A bare-chested picker emerging from the combs of a branch on the ground.

-2

-1

2

1

21

15

A plate containing about fifteen pieces of capped honeycomb.

2

0

2

-1

22

38

A dish containing chihondro (an energy bar made with honey and peanuts).

0

-4

3

1

23

16

A farmer with two children posing in front of two plastic water bottles filled with honey and a white food can. In the background, traditional houses (Madagascar).

-2

1

1

1

24

32

A market stall with a dozen 1 kg plastic jars of honey, a block of wax and a scale.

0

-3

1

3

25

36

A queen with a dot of paint on her thorax. The queen lays eggs surrounded by workers.

1

3

-2

-3

26

33

The flight board of a hive with a large number of foragers

3

-3

0

-1

27

26

Four beekeepers with protective veils and smokers pose in front of hives.

-1

-1

-1

3

28

3

A beekeeper visiting a hive. Using a smoker, he smokes the tops of the frames.

-1

-2

-1

-2

29

12

Around fifty Dadant hives divided into two lines and set up under fruit trees. The land is flat, the grass has been mown, and behind it is a wooded slope.

0

2

3

2

30

4

Closed brood occupies almost the whole of a Dadant frame.

0

1

1

0

31

10

Two labelled 500g jars containing fairly clear, crystallised honey, next to the same jar containing pollen.

0

0

2

2

32

17

Two people stick labels on 1-litre plastic bottles or 500g jars. Inside is dark liquid honey.

0

-3

0

0

33

13

A pile of around thirty dead bees on a hive board

-3

0

-1

0

34

25

A sign in front of beekeepers in overalls reading: "Death of bees=Death of humanity".

-1

1

2

-2

35

1

A dense tropical forest from the sky (Mayotte).

4

4

4

4

36

34

In an agricultural plot with banana trees, the soil is bare, with a pile of dried branches in the foreground. Below the plot, woodland and settlements are visible.

-2

3

3

3

37

37

A plot of forest with felled trees, a dozen people are at work. This plot will soon be cultivated, but the forest in the background looks untouched.

-3

4

0

0

38

39

A plot of farmland with the ground blackened by flames (slash-and-burn farming). The plot borders a path where people are chatting under a tree. In the background, the landscape is wooded.

-4

3

2

-1

39

35

A man in a cap is spraying vegetable crops with a backpack sprayer (insecticide). He is wearing no protection. In front of him, a cloud of water spray is clearly visible.

-4

2

1

-1

Figure 2: The forced distribution diagram

Figure 2: The forced distribution diagram

9The thirty-nine photos (Table 2) were sorted as follows:

  • the respondent was asked to stack the photos he liked on his right and the ones he didn't like on his left. The two piles thus formed were not necessarily the same height;

  • from the pile on the right, the respondent had to choose their ten favourite photos, then from these ten photos, their two favourites, and from the remaining eight, their three favourites. Two columns (+4 and +3) appeared. The five photos still in hand corresponded to the +2 column. At this point in the Q-sort, three of the nine columns had been completed. The photos were placed on the floor out of the respondent's sight, while the other photos in the pile remained on the table to the respondent's right;

Table 2: Factor loadings

Table 2: Factor loadings
  • for the pile on the left, the instructions were exactly the same. They revealed the columns -4, -3, -2, i.e. the ten least liked photos in our sample (or Q sample);

  • On the table, the respondent was now faced with two piles containing a total of nineteen photos. There were three possible scenarios: a) "one of the two piles contained six photos", the +1 or -1 column appeared without any further sorting, and in the remaining pile (thirteen photos), the respondent only had to choose six (favourites in a pile on the right or least favourites in a pile on the left), the remaining seven photos corresponded to the neutral value (0); b) "one of the two piles contained five photos or less", these photos were placed in the +1 or -1 column, and the respondent was asked to choose from among those remaining (between 14 and 18 or even 19 photos) the photos needed to complete this same +1 or -1 column, then the six preferred or least preferred photos to make the -1 or +1 column appear. The remaining seven photos corresponded to the neutral column. c) "Both piles contained at least seven photos each". In the right-hand pile, the participants were asked to choose their six favourite photos, and the remaining photos were placed in the neutral column; in the left-hand pile, the last pile still on the table, the same request was made.

10The Q-sorts were then analysed using the Ken-Q application4. All the calculations are generated automatically by the application, depending on the desired parameters. For the factorial extraction, we selected eight "centroid factors", then four factors for the "Varimax rotation"5. The factor loadings calculated by the Ken-Q application are presented in the form of a table (Figure 3). Column 1 shows the identifiers we assigned to the respondents ("resp1", "resp.2", etc.); columns 3, 5, 7 and 9 show the factor load (or saturation) for each identifier. Note that the closer this load is to the value 1, the more representative the sorting is of the factor in question. In this case, we asked for the highest loading factor to appear in colour, and for the indicator factor to be selected automatically. The Figure 3 shows, among other things, that: the first nine respondents helped to define the four factors (since the four corresponding colours appear); the white box factor applies to half of all respondents; the first respondent is the one with the highest saturation, with 0.752 for "Factor 1"...

Figure 3: Factor matrix

Figure 3: Factor matrix

2.2. Sampling based on pre-selected profiles

11In order not to exclude people who are unfamiliar with the Internet - web applications exist to carry out Q-sorts - we decided to be present and to provide the necessary equipment. The Q-sorts were carried out in the presence of two or three facilitators: one gave the instructions in French and memorised the elements of non-verbal communication; a second facilitator was responsible for recording the sortings on a sheet of paper and photographing the Q-sort; a third translated the instructions in shimaoré or shibushi when necessary and then laid out the photos on the ground one by one.

12We opted for non-probability sampling, as it was better suited to an exploratory approach. In order to take account of socio-professional and/or socio-demographic variables, seven profiles were selected: "religious leaders", "elderly people", "shopkeepers", "honey gatherers", "farmers", "beekeepers", "young people" and "managers of natural areas". It should be noted that honey gathering may be practised by people corresponding to another profile among those defined above. The sampling process was based on the relative heterogeneity of viewpoints, not on a rule of proportionality. Two non-probabilistic methods were used. "Snowball sampling" consists of finding people to interview through networks of acquaintances. We asked people if any of their personal or professional contacts may match the profiles we were seeking. In general, a meeting was arranged at the person's home or place of work. The second sampling method, known as "blind sampling", was used mainly for certain profiles such as "elderly people", "shopkeepers" and "young people". This is about asking random people at the survey site.

13Our sample was made up of thirty-five people (23 men and 12 women) who met in February and March 2023. Twenty-nine of them completed a full Q-sort, i.e. all the photos were arranged in order of preference. The Q-sorts that could not be included in the factor analysis most often concerned very elderly people. The average age of those surveyed was 50.4, with a median age of 45. The youngest was 16, the oldest over 95. Our sample included 23 working people, six of whom were shopkeepers (grocery shops or traditional markets), five farmers (four of whom were beekeepers with a second job), and four civil servants working in departments responsible for the environment in the broadest sense. We also met seven retired people - most of whom were former farmers - four high school students, including two university students on Nature Management, and a young jobseeker (a manual worker).

3. Results

14The analysis will be carried out as follows: we will first describe the five photos with the highest scores, then the ones with the lowest scores; we will provide a description of the sorting factors and, where applicable, the profiles of the individuals who helped define them.

3.1. The top five photos compared to the lowest scores

15The results show a very clear preference for the photo representing ‘A dense forest seen from the sky’ (Table 3). This photo received the maximum score of +4 in half of the Q-sorts. The average score is 2.64. It is also the most consensual photo (Supplementary Table 1). The second most popular photo is of a ‘Honey bee foraging pollen’, with an average of 1.47. In comparison to the previous photo, it only scored +4 on four of the twenty-nine sortings carried out. The scores of the other three favourite photos are quite similar. The fact that they illustrate specific fields implies the multidimensional nature of respondents’ perceptions of pollinating insects (habitats, species, practices, uses). The second favourite insect is Danaus chrysippus, also known as little Monarch (rank 3), and the following insects are all Rhopalocera. On the other hand, for the five least favourite photos, only two fields were considered: the potential danger posed by the honey bees to humans, and conversely, the threats posed by human activities to insects. The photo that showed an allergic reaction to the face after a Hymenoptera sting was given the lowest score (-2.04). More than a quarter of respondents rated this photo -4 eight times. The ‘man treating market vegetable crops’ is like that. This photo had the most diverse score, with a standard deviation of 2.44 (Table 3).

Table 3 : The top five photos compare to the lowest score

Table 3 : The top five photos compare to the lowest score

3.2. The four sorting factors used to characterise the profiles

16Four sorting factors were highlighted by factor analysis based on photographic preferences. Due to the number of photos to sort and the number of themes covered, the cumulative variance explained was 56%, which is a significant amount (Supplementary Table 2). These four factors have been sub-titled to make reading easier (Table 4).

Table 4 : The four factor identified from the factor analysis

Table 4 : The four factor identified from the factor analysis
  • Factor I, which is titled "Pollinating insects threatened by human activities", has an unrivalled attraction for insects, particularly for butterflies. Upon reviewing the twelve favourite photos of the individuals in question, eight of them display a pollinating insect. There are four Rhopalocera present, including the little Monarch (D. chrysippus) with a score of 1.18. The honey bee, which is featured in three of the five favourite photos, including a close-up of a foraging bee (1.85), is not surpassed by the number of butterflies in the ranking. The five favourite photos show orders of undifferentiated insects resting on a flower, suggesting that foraging is a factor of choice for people to define this factor. Often, these species can be easily observed. Pollinating insects, except for moths (Achaea sp. and Lymantriinae sp.), are viewed positively, even those known to inflict painful stings. The wild bees colony that was built under a thick branch is valued (rank 3 out of 29) in Factor I, as we will see in other factors. Numerous bees at the entry of the hive is also very popular. On the other hand, an apiary, a brood frame, honey in jars, and packing in labelled pots seem to leave people indifferent, which is less true for the pieces of combed honeycomb. This may indicate a preference for wild honey. The photos with a negative charge on Factor I display the threats to pollinators, including fire, pesticides, honey harvesting and deforestation. The pile of dead bees is a characteristic photo of Factor I, just like the discarded wax combs. It is demonstrated that certain threats to honey bees have direct consequences. Logically, the profiles of those who identified this factor (12) include the beekeepers (4). There are also engineer and technicians who specialize in agriculture or forestry management. These are men between 20 and 45 years old who have a higher education diploma. The other four profiles exhibit a saturation of the synthetic factor of between 0.37 and 0.53, which makes them less representative. Keep in mind that the ideal sort, which is a sort that is rigorously representative of the factor in question, would have a value of 1.

  • Factor II, titled "The cultivation of the land and the packaging of the honey" is defined by five types with a maximum saturation (or correlation) of 0.66. Compared to the previous factor (0.88), this is less representative. The focus here is on natural or semi-natural environments, and not so much on pollinators. Out of the twelve favourite photos, only three show an insect. The carpenter bee, Xylocopa caffra, was ranked first by respondents (rank 8), followed by the honey bee (10) and a butterfly, Junonia rhadama (12). The other species were perceived as neutral, except for the mango flower beetle and a moth (Lymantriinae sp.), which were perceived negatively. The appreciation for agricultural practices, even when they are detrimental to the entomofauna or suggest that they are, is what stand out here. Tree cutting, for example, is the second most popular photo, with a score greater than 2, making it a distinctive photo. The agricultural plot with banana trees and dry soil also discriminate against this factor positively. With a score of 0.82, the farmer who sprays pesticides on his vegetable crops using a backpack sprayer ranks 7th out of 39. The photo that had the least popularity was the one showing the facial allergy after an insect bite, with a score of -2.28. It is more surprising to find here a traditional culinary preparation from the neighbouring island of Anjouan, chihondro (-1.39). Neither a market stall with a saleswoman in brightly coloured clothes behind a few plastic pots containing honey and a large block of wax, nor the bottling of dark honey according to commercial standards are appreciated. On the face of it, then, it's not just the 'traditional' aspect, or even the perception of a form of archaism, that might be disturbing. The profile of the respondents who define this factor corresponds to the technicians – the two religious leaders we met also present – and above all to the men, aged 35 to 55.

  • Factor III, titled "Honey bees and insect phobia", concerns four people, three men and one woman, all elderly shopkeepers and one retired person. In this case, pollinating insects are the object of undesired fear or disgust. The only one insect that was perceived positively was A. ranavalona, an easily observed butterfly. With a score of 0.32, it was ranked 17th out of 29. The foraging honey bee was ranked 27th with a score of -0.46. The wasp Sphex torridus (-1.68) topped the list of rejected insects, ahead of a moth (Achaea sp.), the carpenter bee (X. caffra) and the mango flower beetle (Protaetia fusca). The choice of the photo showing an allergic reaction (-1.54) tends to support the hypothesis that the evocation of a painful sting, especially from a wasp, would be enough to negatively affect this factor. It’s surprising that the two of the four favourite photos of the respondents are: a wild bees colony built under a branch, and an apiary, given that the bee is also has a vulnating stinger. The combination of the wild and the domestic bees may reflect an acceptance of the local development of modern beekeeping. The score given to the wild colony (1.95) should not be overlooked, as the attachment to honey-gathering practices is obvious, not the wonder that the sight of built combs in the middle of nowhere can evoke. Furthermore, the respondents seem to be able to tell the distinguish between this photo and a naked swarm hanging from a branch, which is perceived as negative. It’s strange that they were unconcerned about a colony installed under the roof of a house, while bees can be a real danger to their occupants. Honey is popular product in the summary sorting, which likely explains the special place of bees and the questions that arise. The packaging of the product seems to be of utmost important to them. They prefer honey in combs or jars with attractive labels to honey sold in plastic jars in the traditional market, or in plastic water bottles. Finally, as for Factor II, chihondro is a distinctive photo, but this time in the positive (1.46). The reason may lie in the family history of the people involved, particularly their connections with Anjouan, as this preparation seems to be deeply rooted in local traditions.

  • Factor IV, titled "Beehive products and their marketing", was defined on the basis of five respondents. Over 65 years old, three of them were often full-time farmers or landowners. Two of the four men have already harvested honey, with one of them being under the age of 20 and the other being over 75. Factor IV has a similarity to the Factor III’s sorting. Both factors are positively discriminated against by the combination of Wild bee colonies (under a branch) and Apiary. The scores were lower for Factor IV were lower: 1.61 versus 1.95 for wild bees, and 0.97 versus 1.53 for the apiary. The tradition market sale of honey and wax is a distinctive photos because it is ranked 3rd in the preferences, whereas it is perceived as either neutral or very negative in the other factors. The ranking of preferences (rank 7) quickly displays another photo of honey and pollen jars, which are glass jars that comply with product labelling standards. Obviously, the respondents who define Factor IV seem to value the marketing of beehive products, particularly honey, and maybe also wax. Their preference is for packaging in jars. Their negative view of the plate is also a result of this preference. The photo is almost unique because the scores for the other factors are still positive: 1.15 for Factor III, 1.00 for Factor I, but 0.04 for Factor II. Neither are the production steps highlighted. Pollinators’ preferences are determined by insect genus or sub-orders for Lepidoptera when they are perceived as positive or neutral. The twelve favourite photos feature only two pollinating insects, Papilio demodocus (0.92) and J. rhadama (0.86). Before rank 15, the other butterflies appear. Then there are honey bees (0.26) and carpenter bees (0.10). Respondents distinguish between species when they are perceived negatively and several species from the same genus or suborder are suggested. Delta emarginatum (-1.72) was rejected more than S. torridus (-0.24) among wasp. The same is true for to moths (-1.57 for Lymantriinae sp. versus -0.13 for Achaea sp.). D. emarginatum, the mason wasp, discriminated negatively against this factor; this was also the case with S. torridus for Factor III. Nevertheless, there is a significant difference. People who define Factor IV are entirely indifferent to the danger posed by insects that are photographed by an allergic reaction. Conversely, throwing brood and honey combs on the ground can cause very negative reactions.

Discussion and conclusion

17Analysis of human interaction with pollinating insects has been a particularly interesting field for research, especially when it is new as it is a Mayotte. In fact, this is what led the organisers of the National Plan for Pollinating Insects to request this social science research. Given the local context and the uncertain level of knowledge in the area under study, the Q-sorts as a preparatory phase was carefully thought through. We selected a fairly large number of photos covering themes that we considered to be priorities, while others were more of an exploratory nature. The results exceeded our expectations. Firstly, because the people we met enjoyed the exercise and took it very seriously; secondly, because it encouraged a more dynamic approach to the discussions, in the knowledge that, depending on the sorting, questions would preferentially focus on selected photos; and thirdly, because a web application made it easy to carry out the factorial analysis that was essential for characterising the sorting profiles. Q-sorts have their limitations when it comes to the relevance of the results they can draw from, even though they are interesting. Despite the large number of photos tending to limit bias, the choice of photos is never neutral. Our perception of insects, for example, can be influenced by macro photography because it can see things that are not visible to the naked eye. In general, it reinforces a positive or negative feeling. When the study site covers a very small area with a relatively poor photographic base, it can also be a challenge to gather a large sample of photos. The use of off-site photos, although perfectly illustrating the themes selected for the study, may give rise to contradictory interpretations and to "errors" in the sorting. Reading a photo may require visual acuity that some respondents may not or no longer have. Once again, bias in sorting can occur. To better understand the choices made, an additional interview remains an essential step.

18The Q-sorts highlighted four profiles, one of which, concerning almost half of the sorting carried out, indicates ecological and environmental sensitivity. This is expressed both by an interest in pollinating insects, particularly hymenopterans and lepidopterans when foraging, and a rejection of human activities when they are perceived as destroying wild fauna and flora, natural habitats (particularly forests), and degrading the living environment. This vision of the environment, which is embodied in the sorting of photos, is generally developed by people with higher education qualifications or who work in land management. In view of the economic, social and demographic reality of Mayotte, this profile is probably a minority, if not a marginal one, in the resident population. This observation would justify the establishment of campaigns to raise public awareness of the role of pollinating insects and the preservation of their environment. In the other profiles defined - they represent three out of five sortings - pollinating insects are often feared, but they can also inspire repugnance. Only butterflies seem to enjoy a positive image in people's minds, on the sole condition that they are not suspected of having been stinging caterpillars or crop pests in their life cycle. The honey bee is fairly well classified, although photographic preferences suggest an interest in honey rather than the insect as such. In the sorting, the packaging of the honey often appears as distinctive, i.e. it helps to define a profile. In contrast to the labelled glass jars of honey on the shelf or in plastic bottles, some people prefer packaging that complies with international trade standards, while more people prefer a product that is perceived as coming from a traditional method. More surprisingly, the photo showing a plate with chihondro, a culinary preparation originating from the neighbouring island of Anjouan, is also divisive. While some people value this preparation made from honey and peanuts, because they associate it with festive meals and family traditions, others see it as being sold on the sly or as a high-calorie food, at least that's what the additional interviews seem to tell us. In fact, there was almost unanimous agreement on a single photo: a dense forest photographed from the sky. The results presented in this article were intended to provide a better understanding of the relationships with pollinating insects, rather than a broader framework that might be the environment, which means that some interesting information could not be processed. They deserve to be.

Haut de page

Bibliographie

Allsopp MH, de Lange WJ, Veldtman R (2008). Valuing Insect Pollination Services with Cost of Replacement. PLoS ONE 3(9); e3128. https://0-doi-org.catalogue.libraries.london.ac.uk/10.1371/journal.pone.0003128

Brown, S.R. 1980. Subjectivité politique : applications de la méthodologie Q en science politique. Londres: Yale University Press.

Chapman A.D. (2009). Numbers of living species in Australia and the world (2nd ed.), A report for the Australian Biological Resources Study; https://www.dcceew.gov.au/science-research/abrs/publications/other/numbers-living-species/discussion-invertebrates#insecta

Dauphin S., Thibault F. (2011). Mayotte : les spécificités du 101e département français, Revue des politiques sociales et familiales, 106: 94-97, https://www.persee.fr/doc/caf_2101-8081_2011_num_106_1_2656

Dieteren C., Patty N., Reckers-Droog V., van Exel J. (2023). Methodological choices in applications of Q methodology: A systematic literature review, Social Sciences & Humanities Open; 7(1), https://0-doi-org.catalogue.libraries.london.ac.uk/10.1016/j.ssaho.2023.100404

French government (2021). Plan national en faveur des insectes pollinisateurs et de la pollinisation 2021-2026; 95p., https://www.ecologie.gouv.fr/sites/default/files/2021.11.21_Plan_pollinisateurs.pdf

Jactel H., Imler J.L., Lambrechts L., Failloux A.B., Lebreton JD, Le Maho Y., Duplessy JC, Cossart P., Grandcolas P. (2020). Insect decline: immediate action is needed. Comptes Rendus. Biologies; 343(3): 267-293, https://comptes-rendus.academie-sciences.fr/biologies/articles/10.5802/crbiol.37/

Klein A.M., Vaissière B. E., Cane J.H., Steffan-Dewenter I., Cunningham S.A., Kremen C. & Tscharntke T. (2007). Importance of pollinators in changing landscapes for world crops, Proc. R. Soc. B.; 274: 303–313, http://0-doi-org.catalogue.libraries.london.ac.uk/10.1098/rspb.2006.3721

Oberlinkels M., Trouillard P., Fouillit V., Boisard M. (2007). L’Atlas des Paysages de Mayotte. Ginger Environnement et Infrastructures ; 241p., https://www.ceb-mayotte.fr/images/4-La_Biodiversite/Le_paysage/2._atlas_paysages_2007.pdf

Perichon S. (2023). Les relations entre la société mahoraise et les insectes pollinisateurs – Analyse d’un recueil de représentations sociales, d’usages et de pratiques à partir de photographies commentées (Q-sort), Rapport d’étude pour la Fédération mahoraise des associations environnementales (FMAE), 70p.

Potts S. et al. (2016). Les pollinisateurs, la pollinisation et la production alimentaire, Rapport d’évaluation, Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES), 35p., https://www.ecologie.gouv.fr/sites/default/files/rapport_evaluation_pollinisateurs-IPBES.pdf

Rost F. (2021). Q-sort methodology: Bridging the divide between qualitative and quantitative. An introduction to an innovative method for psychotherapy research. Couns Psychother Res.; 21: 98–106. https://0-doi-org.catalogue.libraries.london.ac.uk/10.1002/capr.12367

Stephenson W. (1935). Correlating persons instead of tests. Character & Personality, A Quarterly for Psychodiagnostic & Allied Studies, 4: 17-24

Wagner D. L., Grames E. M., Forister M. L., Berenbaum M. R., & Stopak D. (2021). Insect decline in the Anthropocene: Death by a thousand cuts. Proceedings of the National Academy of Sciences of the United States of America, 118(2), [e202398918]. https://0-doi-org.catalogue.libraries.london.ac.uk/10.1073/PNAS.2023989118

Watts S., et Stenner, P. (2012). Doing Q Methodological research: theory, method and interpretation, Sauge Publications Ltd. Doi: 10.4135/978144625191

Haut de page

Notes

1 https://daaf.mayotte.agriculture.gouv.fr/la-foret-a-mayotte-r69.html

2 https://inpn.mnhn.fr/collTerr/outreMer/choix/976

3 https://uicn.fr/liste-rouge-france/

4 https://shawnbanasick.github.io/ken-q-analysis-beta/index.html#section1

5 The Varimax rotation (also known as the Kaiser-Varimax rotation) maximises the sum of the variance of the loadings squared, where "loadings" means correlations between variables and factors.

Haut de page

Table des illustrations

Titre Figure 1. The study site
URL http://0-journals-openedition-org.catalogue.libraries.london.ac.uk/etudescaribeennes/docannexe/image/30308/img-1.png
Fichier image/png, 315k
Titre Figure 2: The forced distribution diagram
URL http://0-journals-openedition-org.catalogue.libraries.london.ac.uk/etudescaribeennes/docannexe/image/30308/img-2.jpg
Fichier image/jpeg, 34k
Titre Table 2: Factor loadings
URL http://0-journals-openedition-org.catalogue.libraries.london.ac.uk/etudescaribeennes/docannexe/image/30308/img-3.jpg
Fichier image/jpeg, 189k
Titre Figure 3: Factor matrix
URL http://0-journals-openedition-org.catalogue.libraries.london.ac.uk/etudescaribeennes/docannexe/image/30308/img-4.jpg
Fichier image/jpeg, 58k
Titre Table 3 : The top five photos compare to the lowest score
URL http://0-journals-openedition-org.catalogue.libraries.london.ac.uk/etudescaribeennes/docannexe/image/30308/img-5.jpg
Fichier image/jpeg, 126k
Titre Table 4 : The four factor identified from the factor analysis
URL http://0-journals-openedition-org.catalogue.libraries.london.ac.uk/etudescaribeennes/docannexe/image/30308/img-6.jpg
Fichier image/jpeg, 73k
Haut de page

Pour citer cet article

Référence électronique

Samuel Perichon, « Using Q-methodology to Understand The Human Interactions with Pollinating Insects in Mayotte (French Island, Indian Ocean of East Africa)  »Études caribéennes [En ligne], 57-58 | Avril-Août 2024, mis en ligne le 30 avril 2024, consulté le 23 juin 2024. URL : http://0-journals-openedition-org.catalogue.libraries.london.ac.uk/etudescaribeennes/30308 ; DOI : https://0-doi-org.catalogue.libraries.london.ac.uk/10.4000/etudescaribeennes.30308

Haut de page

Droits d’auteur

CC-BY-NC-4.0

Le texte seul est utilisable sous licence CC BY-NC 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.

Haut de page
Search OpenEdition Search

You will be redirected to OpenEdition Search