Analysis of Mushroom Consumption and its Determinants in Ghana

The study was carried out in three regions of Ghana; Greater Accra, Ashanti and Bono-East region. The regions were selected because they are the regions where the bulk of mushrooms are produced in the country and, invariably, the regions with the highest number of mushroom consumers. According to data from the Mushroom Growers and Exporters Association of Ghana (MUGREAG), Bono-East region is the largest producer, followed by Greater Accra and Ashanti region. The Greater Accra region is the seat of government and the regional capital city is Accra, which is also the capital of Ghana. The region has 29 districts. It is the most populated region with a population size of 5,455, 692 (GSS, 2021). It is also the most urbanised region in the country, with 87.4% of its population living in urban areas (Songsore, 2009). Ashanti region is the second most populous region, with a population size of 5,440,463 (GSS, 2021). The region is the second most urbanised region in the country after Greater Accra region. The region has a total of 43 districts. Bono East region is one of the newly created regions. The capital of the region is Techiman. The region has a total population size of 1,203,400 (GSS, 2021). The region has eleven districts. Agriculture is the main economic activity in the region. The main crops grown include yam, cassava, cocoyam, beans, rice and plantain. Fig. 1 shows the map of the study areas where the respondents were sampled.

Fig. 1.

Map of the study area

Source: Centre for Remote Sensing and Geographic Information Services, University of Ghana (2022).

The target population were mushroom consumers in Greater Accra, Ashanti and Bono East regions. The data was collected using a well-structured questionnaire through Computer Assisted Personal Interview (CAPI). A Multistage sampling technique was used to select the respondents. First, a purposive sampling of the three regions were made because they are the highest mushroom producing regions in the country and, invariably, with the highest number of consumers too. According to MUGREAG, Bono East region has the highest number of mushroom farmers, followed by Greater Accra region and Ashanti region. Hence, it is expected that a lot of mushroom consumers will be in these regions. Also, Greater Accra and Ashanti regions were selected because they are the most urbanised regions in the country, and studies have noted that urbanisation drives demand for processed foods (Andam et al., 2015). Two approaches were used to select the respondents for the interviews. The first approach was to directly contact mushroom farmers in these regions to get the list of their clients, if available, and select them at random for the interviews. A list of contacts of consumers were obtained from some of the farmers in the three regions. The respondents were selected at random and contacted. A few responded positively, while others did not. Even for those who responded positively, it was difficult booking appointments with most of them for the interviews. In all, less than 10% of the contacts were booked and interviewed.

The second approach was to station enumerators at mushroom sales points, in various local shops and supermarkets, in order to collect contact details of shoppers and book appointments with them for interviews later. Shoprite, in Accra Mall and Koala, as well as ten local markets were selected in the Greater Accra region and visited. The local markets included Malam Ata market, Agbogboloshie market, Makola market, Tema community one market, Kaneshie market, Madina market, Adenta shopping centre, Nima market, Dodowa market and Dome Kwabenya market. Unfortunately, mushrooms were sold in only four local markets (Madina market, Adenta shopping center, Dodowa market and Dome Kwabenya market) at the time of the data being collected. The information we gathered from the other six markets was that the traders were selling only wild mushrooms, and wild mushrooms were out of season at the time of the data collection. Kumasi City Mall as well as six local markets were identified and visited in the Ashanti region. The local markets included Kejetia market, Asafo market, Ayiga market, Bantama market, Suame market and Ejisu market. At the time of the data collection, mushrooms were sold in three (Kejetia market, Asafo market and Ejisu market) out of the six markets. The information gathered was that the traders in the other three markets were selling only wild mushrooms, which were out of season at the time of the data collection. Kejetia market is the biggest market in the Ashanti region, where people from different parts of the region come to shop. In the Bono-East region, two local markets were identified and visited, namely Nkoranza South market and Techiman Municipal market. We could not identify supermarkets where mushrooms were sold. Hence, we were stationed at the local markets only. The snowball sampling technique was also employed to locate other respondents in the region.

The population of mushroom consumers in the three regions are unknown, therefore the standard sample size determination formula, for unknown population, was used to determine the sample size for households. The standard sample size determination formula is specified as in equation (1). (1) n=[Z2⋅p(1−p)]e2 \[n = {{[{Z^2} \cdot p(1 - p)]} \over {{e^2}}}\] where n is the sample size, Z is the confidence level (1.96 for a confidence level (α) of 95%), p is the expected proportion of interest to be studied (50% = 0.5 for unknown population) and e is the percentage error margin (0.05). Based on the formula, the minimum sample size of 384.2 is expected for the study. A total sample size of 550 respondents (Greater Accra 220, Ashanti region 180, Bono-East region 150) was used for the study.

Two types of information are usually gathered from food consumption studies: consumption or not and consumption frequency (Jiang et al., 2017). The study considered only mushroom consumers. Non-consumers of mushrooms were not interviewed. Therefore, the information gathered was mushroom consumption frequency. The consumption frequencies were scaled from zero to three, with a higher number indicating more frequent consumption (i.e., 0 = occasional (at least once a year), 1 = 1–2 times a month, 2 = at least once a week, 3 = more than once a week). The frequency of consumption was analysed using descriptive statistics. Given the categorical nature of the dependent variable i.e., consumption frequency, the factors influencing mushroom consumption frequency could be analysed using ordered probit model or ordered logit model. Ordered probit model was used in this study. Several studies have used the ordered probit model in the analysis of factors influencing consumption frequency (Abusin et al., 2022; Akbay et al., 2007; Asante-Addo and Weible, 2020; Lee and Nam, 2019; Nguyen et al., 2015; Terin, 2019). The latent regression model or random utility model serves as the basis for the ordered probit model (Greene and Hensher, 2009), which is specified as in equation (2). (2) n=[Z2⋅p(1−p)]e2 \[n = {{[{Z^2} \cdot p(1 - p)]} \over {{e^2}}}\] where y* is the unobserved or the latent variable, which describes the underlying continuous consumption of mushrooms. x_i is the selective explanatory variables, β is the coefficient vector and is the error term. The latent variable y* can be observed in a discrete form through a censoring mechanism, which is specified in equations (3) to (6). (3) yi=0 if μ−1<y*≤μ0 \[{y_i} = 0\,{\rm{if }}{\mu _{ - 1}}< y* \le {\mu _0}\] (4) yi=1 if μ0<y*≤μ1 \[{y_i} = 1\,{\rm{if }}{\mu _0}< y* \le {\mu _1}\] (5) yi=2 if μ1<y*≤μ2 \[{y_i} = 2\,{\rm{if }}{\mu _1}< y* \le {\mu _2}\] (6) yi=j if μj−1<y*≤μj \[{y_i} = j\,{\rm{if }}{\mu _{j - 1}}< y* \le {\mu _j}\] where μ₀ < μ₁ < … < μ_j = unknown threshold parameters to be estimated using a sample of n observations. The dependent variable used in the model is y = 0, y = 1, y = 2, y = 3 representing a households mushroom consumption frequency equal to at least once a year, 1–2 times a month, at least once a week and more than once a week, respectively. Following Greene and Hensher, (2010), the probability of a respondent selecting one of the frequencies is given as in equation (7) to (8). (7) Prob(y=0|x)=F(−x′β) \[{\rm{Prob}}(y = 0|x) = F( - x'\beta )\] (8) Prob(y=1|x)=F(μ1−x′β)−F(−x′β) \[{\rm{Prob}}(y = 1|x) = F({\mu _1} - x'\beta ) - F( - x'\beta )\] (9) Prob(y=2|x)=F(μ2−x′β)−F(μ1−x′β) \[{\rm{Prob}}(y = 2|x) = F({\mu _2} - x'\beta ) - F({\mu _1} - x'\beta )\] (10) Prob(y=3|x)=1−F(μ2−x′β) \[{\rm{Prob}}(y = 3|x) = 1 - F({\mu _2} - x'\beta )\] where F (.) is the cumulative distribution of ε_i. For the ordered probit model, F (.) is represented by the cumulative probability function of the univariate normal distribution. The specifications of the likelihood function and log likelihood functions are given in equation (11) and (12), respectively. (11) L=∏i=1N∏j=0NyijFμj−xi'β−Fμj−1−xi'βyij \[L = \prod\limits_{i = 1}^N {\prod\limits_{j = 0}^N {{y_{ij}}\left\{ {F\left( {{\mu _j} - {x_i}'\beta } \right) - F\left( {{\mu _{j - 1}} - {x_i}'\beta } \right)} \right\}} {y_{ij}}} \] (12) LogL=∏i=1N∏j=0NyijlogFμj−xi'β−Fμj−1−xi'β \[LogL = \prod\limits_{i = 1}^N {\prod\limits_{j = 0}^N {{y_{ij}}\log \left\{ {F\left( {{\mu _j} - {x_i}'\beta } \right) - F\left( {{\mu _{j - 1}} - {x_i}'\beta } \right)} \right\}} } \] where y_ij is a binary variable that takes the value of 1 when a respondent i chooses consumption frequency j, and 0 when otherwise. The log-likelihood function in equation (12) is maximised with respect to the parameters of β along with the thresholds μ₀, μ₁, μ₂, μ_j−1, by an iterative procedure to give maximum likelihood estimates (MLEs) of both sets of parameters (Greene and Hensher, 2009). The parameters of β are interpreted as the effect of the explanatory variables on the frequency of consumption (Asante-Addo and Weible, 2020).

Theoretical and empirical studies on factors influencing food consumption revealed that food consumption is influenced by several factors including individual characteristics, product characteristics and institutional or environmental factors (Gama et al., 2018; Pallegedara, 2017; Randall and Sanjur, 1981; Shepherd, 2007). Socio-economic variables that are often included in empirical studies on determinants of food consumption are age, gender, marital status, education, occupation, income and household size. Product characteristics that are often included in empirical studies on determinants of food consumption are price, nutritional value, taste and ease of preparation. Environmental factors that are often included in empirical studies on food consumption are availability, location and degree of urbanisation. Following studies by Boin and Nunes (2018), Oguntoye et al. (2022) and Adegbenjo et al. (2020), the factors hypothesised to influence mushroom consumption frequency in this study included gender, level of education, marital status, age, occupation, income, household size, regional location, availability, taste, price, nutritional value and ease of preparation. The specification of the ordered probit model with the explanatory variables is specified in equation 13. (13) freqmc=β1Age+β2Marstat+β3Gen+β4Edu+β5Incom+β6Occu+β7HHS+β8Loc+β9Avail+β10Taste+β11Price+β12NutriValue+β13EasePre+εi \[\matrix{{{\rm{fre}}{{\rm{q}}_{mc}} = {\beta _1}Age + {\beta _2}Marstat + {\beta _3}Gen + } \cr {{\beta _4}Edu + {\beta _5}Incom + {\beta _6}Occu + {\beta _7}HHS + } \cr {{\beta _8}Loc + {\beta _9}Avail + {\beta _{10}}Taste + {\beta _{11}}\Pr ice + } \cr {{\beta _{12}}NutriValue + {\beta _{13}}Ease\Pr e + {\varepsilon _i}} \cr } \] where freq_mc is the dependent variable representing individual i mushroom consumption frequency, β – are the parameters to be estimated and ε_i – represents the error term. The description and measurement of the individual explanatory variables with their expected signs are presented in Table 1.

The demographic characteristics of respondents are presented in Table 2.

The mean age of respondents is 40 ± 13.14 years. This indicates that the majority of respondents are still active and energetic. The majority of the respondents are females (63.82%). This could result from the fact that women are generally in charge of food preparation in the household. The majority of the respondents are married (64.18%). This indicates that married people consume mushrooms more than those who are not married. The reason for this could be that married people live a healthier lifestyle and consume a more healthy diet than people who are not married. The majority of the respondents have tertiary education (35.27%). This could be as a result of the fact that mushrooms are consumed for its nutritional and health benefits, and it might be the case that higher education enhances nutritional knowledge. The majority of the respondents (60.9%) are employed in the informal sector. The mean monthly income is GHS1129.06±61.57. This means that mushroom consumption is not associated with people with high income levels. The mean household size is 5±4.37. This means that mushroom consumption is associated with smaller household size.

The frequency of mushroom consumption by households is presented in Table 3. The results show that the majority (40.18%) of households consume mushrooms 1–2 times a month. This finding is consistent with Predanócyová et al. (2023), who also found in Slovakia that the majority of consumers consume mushrooms 1–2 times a month. A few households (8.36%) consume mushrooms occasionally, for example, once a year.

The factors influencing a household’s mushroom consumption frequency was analysed using the ordered probit model. The result is presented in Table 4.

Age is significant at the 1% level and is positive, which means that mushroom consumption increases with age. This finding is consistent with previous studies by Boin et al. (2018) and Oguntoye et al. (2022), who reported a similar relationship between age and mushroom consumption. Tertiary education is significant at the 5% level and is positive. This means that higher education enhances nutritional knowledge by identifying mushrooms as a nutritious and healthy food item. This finding is consistent with previous studies by Boin and Nunes (2018) and Oguntoye et al. (2022), who found a similar relationship between education and mushroom consumption. The variables ‘price’, ‘availability’, ‘nutritional value’ and ‘easy to prepare’ are all significant at the 10% level and indicate the perception of importance to mushroom consumption. The coefficient of price is negative, determining a lower likelihood of consuming mushrooms more frequently. The variables ‘availability’, ‘nutritional value’ and ‘easy-to-prepare’ are all positive, determining a higher likelihood of consuming mushrooms more frequently. The coefficient of regional location (Ashanti region) is significant at the 10% level and is negative. This means that respondents in the Ashanti region have a lower likelihood of consuming mushrooms more frequently compared to respondents in the Bono-East region.

The marginal effects of the ordered probit model are presented in Table 5. According to Greene and Hensher (2010), the marginal effects, rather than the estimated coefficients from the ordered probit model, are informative. The marginal effect shows how a change in an explanatory variable affects the expected frequency of mushroom consumption. For ordered probit, estimated marginal effects for each variable across the categories of consumption frequency is not statistically different from zero (Greene and Hensher, 2010). The likelihood that a person will choose at least one category must therefore be weighed against the likelihood that they won’t choose any other categories (Kumar et al., 2008). With increasing consumption, a shift in the marginal effect sign from negative to positive indicates an increase in marginal utility, whereas a shift from positive to negative indicates a decrease in marginal utility (Kumar et al. 2008).

The results of the marginal effects show that a one-year increase in the age of a respondent is negatively associated with a higher likelihood of consuming mushrooms at least once a year and 1–2 times a month. This means that an increase in age increases the likelihood of consuming mushrooms more frequently. This finding could be due to the fact that older people are more health conscious and tend to consume less energy dense foods and more vegetables. This finding is consistent with previous studies by Boin et al. (2018) and Oguntoye et al. (2022), who found a positive relationship between age and mushroom consumption frequency. Tertiary education is negatively associated with a higher likelihood of consuming mushrooms at least once a year and 1–2 times a month. This means that respondents with tertiary education have a higher likelihood of consuming mushrooms more frequently. This finding could be due to the fact that most people consume mushrooms for their nutritional and health benefits and higher education enriches nutritional knowledge. This finding is consistent with previous studies by Boin and Nunes (2018) and Oguntoye et al. (2022), who also found a positive relationship between education and frequency of mushroom consumption.

Price is perceived as being negatively associated with a higher likelihood of consuming mushrooms more than once a week and at least once a week. This finding could be due to the fact that most people perceive mushrooms to be expensive relative to its close substitutes like meat and fish. This finding is consistent with a previous study by Jiang et al. (2017), who reported that mushrooms with a reasonable price increases consumption frequency. Perceptions of ‘availability’ and ‘nutritional value’ are all positively associated with a higher likelihood of consuming mushrooms more than once a week and at least once a week. This finding is consistent with previous findings by Oguntoye et al. (2022). The perception of ‘easy-to-prepare’ is negatively associated with a higher likelihood of consuming mushrooms at least once a year and 1-2 times a month. This means that a perception of ‘easy-to-prepare’ increases the likelihood of consuming mushrooms more frequently. This finding is consistent with previous findings by Oh et al. (2002). Regional location (Ashanti region) is negatively associated with a higher likelihood of consuming mushrooms more than once a week and at least once a week, compared to respondents in Bono-East region, which is a major mushroom producing area.