The methodological approach revolves around the collection of data, their processing and the analysis of the results. The collection method used for the survey is the semi-structured door-to-door interview [11] . The interviewer goes to each household, asks questions to the Head of Household and writes down the answers using a smartphone equipped with the KoBoCollect application [12] .

A pre-survey was carried out and served as a basis for the formulation of certain hypotheses. These hypotheses were verified at the end of the actual survey by inferential statistics.

The survey took place in the community of Djougou in northern Benin (Figure 1) four (04) districts were drawn by lot. By district, four (04) districts were again drawn, which leads to a total of 16 districts namely: Petoni Poho, Zongo, Taїfa, Kilir, Zountori, Batoulou, Formagazi, Baparape, Partago, Tepredjissi, vanhoui , Momongou, Bougou 1, Bougou 2, Kpandouga and Kpaouya.

The survey took into account socio-demographic characteristics in relation to household cereal consumption habits.

The sample size in each survey area is calculated by the method of [13] .

X = Z α 2 p ( 1 − p ) i 2 a v e c p = n N

N: Total number of households in the study area

n: Total number of households in selected neighborhoods or villages

i: 5% = 0.05

Z_α = 1.96

Considering the demographic data from the general population and housing census (RGPH4) carried out by INSAE in 2013, it comes out that the commune of Djougou has 12 arrondissements and a total of 34,039 households. Out of these 12 arrondissements, 4 have been chosen at random for the survey. These are the arrondissements of Djougou 1 with 5230 households, Djougou 3 with 3488 households, Partago with 3723 households and Bougou with 1330 households. These households add up to: 13,771. Hence, we have:

· Total number of households in Djougou: N = 34,039

· Total number of households in the selected area: n = 13,771

Therefore, p = n/N and we get: p = 0.4

At this step, we apply the sampling method (Schwartz, 1995) to obtain the number of households to survey. Let X be that number. We get:

X = ( ( Z α ) 2 ⋅ p ⋅ ( 1 − p ) ) / ( i ) 2       = 1.96 × 1.96 × 0.4 × 0.6 / ( 0.05 × 0.05 )       = 368.7936

We have to round up and we obtain X = 369 households to survey.

As a prelude to the actual survey, a pre-survey was carried out on a sample of 100 randomly selected households in the town of Djougou. The data and results of the pre-survey made it possible to formulate a certain number of hypotheses which were subsequently tested.

With regard to the investigation itself, three (03) major stages have marked its progress.

The first step consisted in the practical training of the investigators. During this training, they received useful knowledge for the smooth running of activities, among other things, the recognition of the geographical limits of the different city districts thanks to counting maps, knowledge and use of the KoBoCollect mobile application. Interviewers also learned how to effectively use the survey questionnaire, which is integrated into the mobile application in the form of an electronic form.

The second stage is a full-scale test phase. Each pair of interviewers was asked to interview five (05) households in their survey areas. This test enabled the management team to assess the average interview time, the level of assimilation of the knowledge acquired and to find solutions to the difficulties reported by the interviewers, in particular the resistance of certain heads of households to answer to the questions. This trial phase made it possible to set the average number of daily interviews at 10 households. In addition, the interviewers all demonstrated a good understanding of the use of the collection tool (KoBoCollect). In view of the results obtained, the test phase was validated, and the third phase was opened.

The third step consisted in collecting the actual data in accordance with the electronic form registered in the application. An average of 10 households per day was set per pair of interviewers. In the event of insufficient Collection, the geographical proximity method has been recommended to compensate for the collection. In each concession, only one head of household was interviewed. In the case of a building, each apartment is considered a house. These conditions were respected to ensure consistent representativeness. The interviews were conducted in Dendi, Yom, Lokpa, Ditamari language and in French.

At the end of the survey, the various data collected were pre-processed by the KoBoCollect application. Further processing was performed with hypothesis testing. The Spyder development environment was used to perform all the necessary calculations, particularly the p-value calculation. The steps for calculating the p-value [14] are as follows:

Step 1: Calculate the probability of success (p) given the data.

p = n N

n: number of positive responses obtained

N: number of people interviewed

Step 2: Calculate the minimum number of successes to expect under the Null Hypothesis.

k = N ∗ H 0

H₀: decimal value of the null hypothesis (0.9 For H₀ = 90%)

N: number of people interviewed

Step 3: Calculate the p-value

p v a l u e = ∑ i = k N C N i p i ( 1 − p ) N − i

The households surveyed consume maize and its derivatives at 96.21% (Figure 7). Our null hypothesis (H₀ = 95% or more) is validated (p = 0.887) and makes maize the first most consumed cereal in Djougou. Similar results have been found by many other researchers [6] [15] [16] , who have jointly drawn the conclusion that maize (Zea mays L.) is a cereal of great dietary importance for many populations in West Africa, mainly in Benin where it constitutes the basis of the diet of the populations of the South and the Center of the Country.

Households surveyed in Djougou consume 59.62% of millet and its derivatives (Figure 7). Our null hypothesis is validated (H₀ = 58%, p = 0.755). Millet is then the second most consumed cereal by the populations in Djougou. [16] Found concordant results by reporting that in the departments of Atacora, Donga and Borgou, millet consumption is in the range of 11 - 18 kg per adult-equivalent/ year and that in the 8 other departments of Benin as well as in large cities, millet consumption is insignificant or non-existent. A study carried out by the CePED in 2010 [17] also found results similar to ours. The populations in the North of Benin have less Western food habits so that certain foods based on millet (“Foura” commonly called “Dèguè” and other millet derivatives are very popular). Outside Benin, millet is also considered a noble cereal because it is often served during traditional ceremonies [18] . This is the case in Burkina Faso, for example, where the majority of food processors (70%) use millet in proportions of 1 and 5 kg to produce the zoom-koom drink [19] .

Figure 7 also reveals that rice is the third most consumed cereal by the populations of the community of Djougou. Indeed, 56.64% of surveyed households consume rice and its derivatives, validating our hypothesis (H₀ = 55%, p = 0.753). Regarding the consump

tion rates of rice and its derivatives, many other researchers have also found similar proportions in Benin and Africa. Indeed, it is reported that strong urbanization, access to the labor market and growing incomes lead city dwellers to consume more imported foods, rice in particular [15] [17] . According to the estimates of [10] , in Senegal in urban areas rice consumption is 77% while in rural areas it represents 59% of cereal consumption. Senegal consumes, on average, more than one million tons of white rice per year [10] .

Sorghum is the fourth most consumed cereal in Djougou, because more than 15.45% of the population of Djougou consumes Sorghum and its derivatives, validating our hypothesis (H₀ = 15%, p = 0.635). Outside Benin, sorghum is just as well consumed in the West African sub-region as it is used in the manufacture of certain local drinks and porridges. In Burkina-Faso, for example, sorghum is consumed even more and it represents 40% to 50% of Burkinabe national production [20] .

In the South of Benin, the populations have a more Western habit so that wheat-based foods such as bread, cake, croissant, pizza, sharwama and others are widely consumed on a daily basis but this is not the case in northern Benin and it is even quite the opposite. Indeed, only 6.5% of surveyed households consume wheat and its derivatives in Djougou with H₀ = 5% and p = 0.920 (Figure 7).

“Digitaria exilis” fonio is a cereal grown in some Sudanian areas of West Africa. Less than 5% of the population in Djougou consumes Fonio and its derivatives. It is a cereal hardly consumed in Djougou and in Benin in general. According to [21] , the very difficult and painful culinary preparation of fonio has made it a marginal cereal in consumption, particularly in towns, and has raised fears of its disappearance. In Guinea, the traditional nature of fonio and the attachment to domestic know-how in its processing explain the limited use of this type of product. In Mali and Burkina, fonio is considered more as an African product of diversification and is less the support of the rural identity of city dwellers [21] .

Table 1 showed that 53.93% of households surveyed get their supplies by purchase, thus validating our hypothesis (H₀ = 50%, p = 0.947). Purchasing is therefore the primary means of supplying cereals in Djougou. Concordant results have been published by other researchers. Indeed, it is reported that purchase is the characteristic mode of acquisition of households in Benin and in Cotonou in particular where it represents 50.7% of supply methods [20] . In addition, purchase is the essential mode of cereal acquisition and represents 51% of expenditure according to [21] . Production (self-consumption) and donations are other forms of cereal supply in Djougou. Indeed, 30.08% of the households surveyed received cereals by donation thus validating our hypothesis (H₀ = 25%, p = 0.988). The donation is the second mode of cereal supply in Djougou. Lourme Ruiz (2017) [22] in his survey found lower rates namely: 10% of cereal donations in Cotonou and 9.3% of donations in Ouagadougou. This same author estimates that the same phenomenon is observed in Dakar. The higher proportions of donations observed in Djougou could be explained by strong socialization within this population.

Table 1 showed that 20.6% of households surveyed produce their own cereals.

Our null hypothesis H₀ = 25% is rejected.

In Djougou, less than 25% of the population is engaged in the production of cereals for self-consumption. [9] Reports in his communication on the supply and distribution of food in French-speaking African cities that: “self-consumption still remains one of the modes of supply in certain cities such as Ouagadougou and Cotonou we meet populations who have an agricultural production activity”. In addition, in Ouagadougou, the survey of [22] estimates at 5.4% the rate of households that produce their own cereals.

In the cereal distribution chain in Djougou, the main suppliers identified are: retailers, wholesalers, producers and supermarkets. The results of our survey reveal that 61.79% of surveyed households get their supplies from retailers. This result validates our hypothesis (H₀ = 55%, p = 0.997). Retailers are therefore the main suppliers of cereals in Djougou (Figure 8). This result can be explained by the fact that retailers are closer to consumers and make their products more accessible. In addition, consumers only need small quantities of cereals to meet their daily needs.

The results of our survey also reveal that 33.6% of surveyed households get their supplies from producers. This result validates our hypothesis (H₀ = 30%, p = 0.933). Producers are the second cereal suppliers in Djougou (Figure 8). This result is explained by the fact that producers act as wholesalers in Djougou by supplying their products to retailers but also to other wholesalers who intend to resell these products in southern Benin and on other local and sub-regional markets. Wholesalers are almost non-existent in Djougou due to the presence of producers. Less than 3% of the population in Djougou get their supplies from supermarkets due to the distribution role that wholesalers play.

Figure 9 shows that 48.51% of households surveyed get their supplies from local markets, thus validating our hypothesis (H₀ = 45%, p = 0.920). This observation is explained by the fact that local markets are made up of retailers for the most part and that they are closer to consumers. [23] In his communication for the benefit of the FAO went in the same direction as us by finding that the purchase is the first mode of supply in cereals and that the rural markets are the first places of purchase. He goes on to say that 17% of households in Cotonou get their supplies from rural markets. This rate is much higher in Djougou in view of our results due to the absence of major commercial infrastructures.

Some households surveyed (31.71%) get their supplies in the streets and alleys, thus validating our hypothesis (H₀ = 30%, p = 0.765). The lack of market infrastructure close to the population was still the cause of the high rate of grain acquisitions in streets and alleys.

Figure 10 revealed that 75.61% of households in Djougou dry cereals at room temperature, thus validating our hypothesis (H₀ = 70%, p = 0.995). Drying at room temperature is the primary technique for preserving cereals in Djougou. Cereal storage in Benin has not undergone significant modernization. In addition, the abundant availability of solar energy favors the technique of preserving cereals by drying at room temperature. In Djougou, 4.07% of households surveyed preserve cereals by protecting them from light and this same proportion of households uses chemical preservatives (Figure 10).

Whether by placing the product away from light or by using chemical preservatives, our hypothesis (H₀ = 5%, p = 0.248) is validated. However, the validation threshold (5%) as well as the low p-value suggests that these techniques are hardly used. The same is true for the technique of preservation by drying in a controlled enclosure used by approximately 3.52% of the households surveyed (Figure 10).

In view of these results, it should be noted that the main technique for preserving cereals in Djougou is drying. [24] [25] [26] also consider drying as the main technique for preserving cereals. As an example, Figure 11 shows field-dried maize.

Reading Table 2 revealed that 51.76% of the households surveyed dry the cereals on site in the field, which validates our hypothesis (H₀ = 50%, p = 0.783). The availability of vast expanses of land and the sun on the one hand and the absence of domestic animals in the field on the other hand favor the drying of cereals in the areas laid out in the field.

About 40.92% of households surveyed dry in areas equipped at home (Figure 12), validating our hypothesis (H₀ = 40%, p = 0.643). The areas set up at home are the second place for drying cereals in Djougou. This result is explained by the availability of space, the ease of monitoring and storing the dried products in the storage places. Houansou et al. (2020) and Rochat et al. (2013) mentioned the importance of drying in areas fitted out at home. The image below shows home-dried maize.

Very few households in Djougou dry their cereals near roads (Figure 13). This is due to the availability of drying space in the field or at home. However, some households dry their cereals in appropriate stores, but these are often owned by agricultural cooperatives (Figure 14).

About 61.79% of households in Djougou store cereals in the kitchen (Figure 15), validating our hypothesis (H₀ = 60%, p = 0.790). The kitchen is used as a place for preparing meals but also as a privileged place for storing foodstuffs, including cereals.

The results show that 20.87% of the households surveyed store cereals in the bedroom, validating the hypothesis (H₀ = 20%, p = 0.669).

About 14.91% of households surveyed store cereals in the store (Figure 15) validating the hypothesis (H₀ = 15%, p = 0.522). The storage of cereals in the living room is observed in 14.09% of the households surveyed.

In view of these results, it was found that the bedroom, the store and the living room are in this order the alternative places for the storage of cereals. These results are explained by the fact that the quantities of cereals acquired by households are not so important to require storage in large infrastructures.

The use of the attic as storage infrastructure is observed in Djougou by 6.5% of the households surveyed. The hypothesis (H₀ = 5%, p = 0.920) is validated, however, granaries are used very little by households in the commune of Djougou (Figure 16). On the other hand, in Atacora in the northwest of Benin, the “banco” type granary used by 69% of households is the most common storage structure [27] .

Table 3 shows that between 2020 and 2021, there was a significant increase in the average cost of all the cereals considered by our study. Other Institutions have also observed these increases. This is the case of the National Institute of Statistics and Demography (INStaD) which in its May 2021 publication reveals a 3.3% increase in the producer price index for food products in the 1st quarter of the year 2021 [28] . By comparing to the same period of the year 2020, the INStAD specified that the producer price index increased by 25.2% and that this evolution is linked to the increase in the cereal indices. (+47.3%) [28] . The price increases thus observed are partly explained by the supply and distribution difficulties due to the COVID-19 pandemic, which paralyzed the freight transport network in general. In addition, the difficulties of access to fertilizers specific to cereal production also explain this price increase. Finally, generalized inflation has contributed to the increase in the prices of basic necessities, including cereals. In addition, INSAE reports in August 2021 that cereal prices have increased in Benin during the last week. She argues that maize prices have increased in the cities of Porto Novo (South-East), Parakou (North-East), Natitingou (North-West) and Lokossa (South-West), while in Cotonou and in Bohicon, prices have fallen, explaining this price variation by the supply of markets.