Motivational Factors Towards Fast-Food Joint Selection in Under-Developed Country Setting: A Partial Least Square and Structural Equation Modeling (PLS-SEM) Approach

Abstract The abrupt rise in the fast food business the world over calls for research attention to the phenomenon, especially, in underdeveloped and developing economies. Research is scanty regarding the phenomenon; especially what motivates patrons to select fast food joints in under developed economies such as Ghana. The study sought to ascertain the motivational factors that actuate (or stimulate) consumers’ intent to select a fast-food joint in an under-developed country setting, particularly, in Ghana, a sub-Sahara African region. Additionally, the partial goal of this survey is to examine the mediating role of convenience (CONV), and taste and preference (TASPRE) given the indirect effect of traditional advertising communication medium (ACM); Radio/Tv and word-of-mouth. Using a quantitative research approach, a structured survey questionnaire was used to intercept buyers of fast-food at vantage points in the Cape Coast metropolis in the Central region of Ghana. A non-randomized sampling technique, precisely, the convenience sampling, was adopted to consider popular fast-food joints that aided the researchers to intercept customers/buyers for the study. Results from the application of partial least square and structural equation modelling (PL-SEM) of 305 valid responses revealed that the mediation (indirect) analysis supported all the mediate-hypotheses. The research implications and future study directions are discussed in the concluding part of the paper.


PUBLIC INTEREST STATEMENT
The fast-food industry continues to grow across developed and developing economies. Yet, the existing literature on fast-food consumption and fast-food joint patronage in several jurisdictions is barely known empirically regarding influential factors in choosing a fast-food joint in a developing country like Ghana, a sub-Saharan African region. After a careful review of literature, we argue that examining the constructs such as desire, taste and preference, convenience, menu price, time, word-of-mouth (WoM) and radio and television advertisements would reveal reasons why patrons in Ghana select fast-food joints. This study would inform practitioners (fast-food vendors) to re-position and re-strategize their marketing plans to lure customers so as to ensure a sustainable fast-food business. Evidence of results and implications of this research is contained in the entire manuscript for readers and fast-foods business operators.

Introduction
Owing to the current dispensation's competitiveness regarding the marketplace, patrons of food products are bombarded with a cornucopia of options, leaving them with the power of choice. Due to this, food vendors, and in this case fast-food vendors need to employ strategies that are customeroriented and customer-friendly to be able to survive. To be able to do this well, fast food vendors must be abreast of the motivational factors driving patrons' fast-food joint selection. Customers may consider a plethora of factors in selecting a fast food joint. Food is a necessary component of human existence, and it comes in several forms and categories. Food provides nutrients and nutrients give the human body energy for several functions such as growth. Food may be prepared and enjoyed at home or bought from food vendors in the restaurants, local chop bars/eateries, hotels or fast food joints. Food joints have been long classified by experts and scholars. One of the fastest-growing joints where people access their ready-to-eat food is what is known as a fast-food joint. In spite of the growing health concerns espoused by nutritionists about fast-foods, patronage of fast-foods keeps rising vis-à-vis the multiplication of fast food joints. This development has brought about a lot of discussion around the fast-food phenomenon. For instance, a report published by DW.com attributes the increasing nature of obesity problems in Ghana to the upsurge in fast-food joints (DW.com, 2019). The fast-food discourse cannot be exhausted due to several factors, including the busy nature of lifestyles coupled with the innovativeness of vendors as well as the rapidity with which fast food joints spring up (Akbay et al., 2007;Park, 2004).
The fast-food discussion has been approached from different angles with different lenses. For instance, fast-food restaurant satisfaction (Qu, 2017); restaurant and food shopping selections (Ayala et al., 2005) and restaurant food healthiness (Kim et al., 2013). Many scholars have conducted studies from different perspectives, using different constructs from different geographical locations and have emerged with different findings and conclusions. For example, Omari et al. (2013) found that 68% of restaurants in the Greater Accra Region of Ghana constitute fastfood joints. This assertion is corroborated by Peters and Kontor-kwateng (2016) who opine that although fast-food is expensive in Ghana as compared to developed countries, it is the fastestgrowing business. Other studies, for example, Amuquandoh and Asafo-Adjei (2013) analysed Ghanaian traditional foods patronised by tourists and found that seventeen local foods topped the list of local Ghanaian foods enjoyed by tourists who visit Ghana.
Fast-food joints keep increasing, especially, in the cities and towns of Ghana (Omari & Frempong, 2016). In the same vein, fast-food consumption keeps increasing. Food and Agriculture Organization (FAO) of the United Nations in a report reveals that there has been a rapid multiplication of fast food activities regarding vending and consumption in Ghana (FAO, 2016). This means that more people are patronizing fast-foods and fast-food joints in Ghana, baring safety, nutritional as well as management concerns. Certain factors may be drawing patrons of fast-foods to the various fast-food joints in Ghana.
Fast-food joints are patronised by many for several reasons. Some of these reasons include convenience, the taste of food, location, the popularity of a fast-food joint, menu price (affordability), advertisements and recommendations from others (word-of-mouth). Prior studies have considered some of these reasons separately and in different jurisdictions. In India, Goyal and Singh (2007) explored the relevance of factors influencing the choice of fast food joint among young Indians and found that taste and quality of food (nutritional values), as well as the atmosphere of the environment of a fast-food joint coupled with hygienic conditions, were the most considered factors in making a choice of a fast-food joint. Similarly, in Australia, Dunn et al. (2011) examined variables that influence fast food consumption and found that taste, satisfaction, and convenience were the major factors influencing fast food consumption. In Ghana, Adam et al. (2014) investigated the relationship between food safety concerns and choice of eating place, drawing a sample from the University of Cape Coast and found that students clothed with food safety concerns preferred eating from restaurants to other eating places in spite of high cost of food in the restaurants. Their (Adam et al., 2014) findings are in sharp contrast to Glanz et al. (1997), who found that concerns regarding nutrition ranked minimum among factors that influence the choice of fast-food and fast-food joints in America. However, the finding of Adam et al. is supported by Pettijohn et al. (1997). Adam et al., however, did not consider the constructs that we intend to investigate. Again, their study considered a limited sample-university of Cape Coast students. Similarly, Kang et al. (2015) found that consumers' patronage of food products is highly prompted by health values. As a point of departure, our survey covers respondents from the entire Cape Coast Metropolis. The kind(s) of food people eat as well as where they buy food is influenced by a barrage of factors, such as proximity and convenience, the time factor, menu price, word of mouth (WoM), taste and preference and radio and television commercials. Based on this, Shepherd's (1999) study explored the Theory of Planned Behaviour (TPB) (Ajzen, 1991) model's efficacy on predicting people's food choice hinged on people's attitudes and beliefs. Once people's food choice can be modelled by their beliefs and attitudes, as predicted by TPB, it is necessary to also examine what influences people's choice regarding food joints. Since one of the fastest-growing phenomena in this regard is fast-food, it is expedient to investigate what influences people to choose one fast-food joint over another, and also investigate the reason(s) for the choice(s). From the foregoing, we observe that most scholars have concentrated much attention on the determinants of food choice and not much has been done on factors that influence customers' choice of fast-food joints in Ghana. In Australia, Thornton et al. (2009) found that buying a fast-food depended on the variety of fast-food joints available.
In spite of the existing literature on fast-food consumption and fast-food joint patronage in other jurisdictions, little is known empirically about factors that influence fast-food patrons in choosing fast-food joints in a developing country like Ghana. We argue that examining the constructs such as desire, taste and preference, convenience, menu price, time, word-of-mouth (WoM) and radio and television advertisements would reveal reasons why patrons in Ghana select fast-food joints. This study would inform practitioners (fast-food vendors) to re-position and re-strategize their marketing plans to lure customers. Whereas in theory, it would help future researchers to examine which of the construct's customers consider most in selecting a fast-food joint not only in Ghana but other developing economies and beyond. To sum up, the main objective of our study, therefore, is to investigate customers' motivation factors in selecting fast-food joints in an under-developed country setting, particularly, in Ghana, a sub-Saharan African region. Again, the partial goal of this survey is to examine the mediating role of research constructs: convenience (CONV), and taste and preference (TASPRE) given the indirect effect of traditional advertising communication medium (ACM); Radio/TV and word-of-mouth (WoM). Using the Partial Least Square Structural Equation Modelling (PLS-SEM), our research question formulated to guide the study is: What are the motivating factors of customers who patronise fast-food joints in Ghana?

Literature review
Fast-food business appears to be on the ascendancy in Ghana (FAO, 2016), and patronage seems to correlate with the rapid nature of the multiplication of fast-food joints in the country (Omari & Frempong, 2016). The popularity of fast-food joints in Ghana cannot be overemphasized. The emergence of the fast-food business in Ghana has changed how and where people eat. According to Peters and Kontor-kwateng (2016), the food business is the greatest and fastest-growing industry in Ghana. They assert that compared to developed countries, fast-food is expensive in Ghana owing to factors such as overhead costs and huge investment coupled with the cachet associated with the fast-food business. This notwithstanding, however, there has been an increase in the number of fast food-joints in Ghana over the years (Mensah et al., 2002).
Factors considered by customers in making a choice of fast food joint may differ from one geographical location to another. In this study, the authors refer to the coastal areas or venues in Ghana where already prepared food is available for purchase, especially, food that can be quickly prepared and eaten at the fast-food joint. The authors, therefore, review the literature on the following constructs: the desire to select fast food joint, taste and preference, convenience, menu price, time, word-of-mouth and radio and television advertisements.

Desire to select fast food joint
Considering the increasing nature of fast joints in Ghana coupled with the increase in fast food patronage, there must be a desire to select one fast-food joint over another. Some scholars have conducted studies on the construct of desire regarding food choice. For instance, Crawford Solberg et al. (2002) tested factors that predict satisfaction by analysing the differences between conflicting facts about what people have and what they desire and found that people's satisfaction depended on the attainment of their desires. Herman et al. (2019) assert that the desire to consume food in a manner that is congruent with the norms of society regarding what is considered suitable for eating, goads people, especially, women's food choice.

Taste and preference
Regarding taste and preference as a determinant of choice of food-food joint, Prabhavathi et al. (2014) found the taste to be the top constituent factor influencing young Indians in patronizing fast-food joints. Taste of food was one of the strongest values considered by customers in patronising fast foods in Korea (Park, 2004). In Ghana, a study conducted by Kwadzo et al. (2013) in Greater Accra, Tema and Kumasi Metropolis found that consumers of meat products in Ghana considered taste among other factors to be one of the factors driving purchase decisions. Akbay et al. (2007) found that preference is one of the factors that influence fast food consumption in Turkey. Goyal and Singh (2007) found that although the young Indian prefers homemade food to fast-foods, in considering food from fast food joints, they consider the taste and quality of food. Also, in the study conducted in America by Glanz et al. (1997) revealed that taste of food was considered by fast-food consumers in selecting fast food joints, and this finding agrees with the finding of Anand (2011). Furthermore, it is also revealing that factors that highly motivate customers' food joint selection in America are taste and preference (Min & Min, 2011).

Convenience
In examining customers' reasons for selecting fast-food joints, convenience appears to be one of the leading factors. In a study conducted by Prabhavathi et al. (2014) to analyse reasons given by customers in India in patronising fast food and fast food joints, they found that convenience ranked second (15%) after taste of food. Urala and Liisa (2003) and Dunn et al. (2011) found convenience to be one of the reasons influencing customers' consumption of fast foods. Convenience as a construct has been found to be one of the major factors that drive the patronage of fast foods and fast food joints, especially in developed countries (Jekanowski et al., 2001). Anand (2011) opines that even though convenience ranks high among factors determining fast food joint selection for dual-income families in India, it is no longer a strong unique selling proposition (USP) for food marketers, and this is also supported by Pettijohn et al. (1997).

Menu price
It appears that not much literature supports the cost of food as an influencing factor in selecting a fast food joint. This notwithstanding, some studies considered menu price as a construct in selecting fast food joints. High menu prices do not deter students' fast food consumption in Ghana (Adam et al., 2014). Interestingly, this finding is incongruent to Glanz et al. (1997), who found that aside from taste, the cost of food was the second most influencing factor in selecting fast food and fast food joints. However, Qu (2017), in a study to discover factors that customers who patronize Chinese restaurants consider important, price and value of food ranked third after "food and environment" and "service and courtesy". He, therefore, concludes that customers in Indiana consider multiple dimensions when making decisions on whether to make a repeat call to a Chinese restaurant. Despite gender influencing food and food joint choice, the cost of food plays a significant role in selecting a fast food joint (Boek et al., 2012).

Time
One of the attributes scholars have tested regarding fast food and restaurant patronage is the time factor. For instance, Harrington et al. (2011) found time not to be too influential in determining consumers' food choices. However, in a study by Kilinc et al. (2013) to identify factors that influence students to select restaurants in Turkey, time factor was the second most pressing factor.

Word-of-mouth
Most patrons of fast food joints make their decisions based on a barrage of factors including recommendations from peers, friends, role models, celebrities and other influencers. Most people visit food joints they had hitherto not visited before but based on recommendations from their friends because they take their friends' words seriously (Krook, n.d.). Krook's findings also revealed that most people prefer new fast food joints to making a repeat call to the same place. A similar study (Kim et al., 2009) found that when customer service is improved, it encourages positive word-of-mouth which eventually increases patronage and strengthens customer loyalty. Word of mouth appears to be the main factor propelling the success of fast food businesses (Basri et al., 2016), even though this also depends largely on other factors such as the quality of the environment in which the food joint is situated.

Radio and television advertisements
Advertising in general, plays a major role in inducing customers' emotional responses (Hyun et al., 2011). Radio and Television (henceforth TV) advertising continue to surge despite the advent of the internet and its related forms of new media advertising. More so, there is a firm relationship between radio and TV advertising and customers' food preferences and purchase decisions (Effertz & Berg, 2017). Singam et al. (2014) aver that advertising greatly influences fast food acceptability by customers. In investigating the performance of radio advertisement on customers' ability to remember product advertisements, Sullivan (2020) concludes that a highly involving music format radio advertisement produces systematically pleasing results.

Summary of research hypotheses and conceptual model
With respect to our literature review, we deduce the following propositions for the research aim ( Figure 1): Hypothesis (H1): Convivence of fast-food accessibility positively and significantly predict customers' desire to select a fast-food joint.
Hypothesis (H2): Time spent to acquire fast-food directly and significantly predict customers' desire to select a fast-food joint.
Hypothesis (H3): The menu price of various fast-foods positively and significantly predict the desire to select a fast-food joint.
Hypothesis (H4): Taste and preference for fast-food positively and significantly predict the selection of a fast-food joint by customers.
Hypothesis (H5): Radio and television adverts potentially predict the customer taste and preference of fast-food.
Hypothesis (H6): Radio and television adverts indirectly and significantly predict the customers' desire to select a fast-food joint through a mediation effect of convenience.
Hypothesis (H7): Word-of-mouth (WoM) regarding the desire for fast-food indirectly and significantly predicts customers' choice for fast-food joint via the mediation effect of taste and preference.

Sample data and demographics
The authors adopted a quantitative inquiry to implement the research objective. To test our research model and the hypotheses, the authors used a survey-based research design to collect quantitative data from a sizeable number of respondents (fast-food consumers) through the intercept approach. The questionnaire was purposefully targeted at consumers of selected popular fast-food joints in the Cape Coast metropolis in the central region of Ghana. A non-randomized sampling technique, precisely, the convenience sampling, was adopted to select popular fast-food joints that aided the researchers to intercept customers/buyers at the time of data collection. A convenience sample is a type of non-probability sampling method where the sample is taken from a group of people easy to contact or to reach. For example, standing at a mall or a fast-food joint (in our case) and asking people to answer questions would be an example of a convenience sample. According to Etikan, Musa, and Alkassim (2016), the technique was adopted as a result of respondents' accessibility, geographical proximity, willingness to participate, participants' accessibility to the researcher, and affordability in terms of the cost associated with reaching the unit of analysis.
Data were finally collected through the intercept approach and online survey by a structured questionnaire. With the intercept, we mean customers were trapped at the fast-food joints while  Source: Authors' own the online survey link was sent to participants who requested it mainly because they were not ready at the time we intercepted them at the designated points. Out of the 400 questionnaires distributed, 371(92%) were returned, while 305 out of the 371 (82%) cases were fully completed and eligible for analysis. Survey respondents were pre-qualified to ensure that their knowledge of fast-food joint selection as well as their level of understanding of motivation factors influencing their selection of s desired fast-food joint.
Data collection was undertaken in the months between November to December 2019. On average, the questionnaire took 5 min to fill. To ensure the ethical standard of our study, respondents were assured of their anonymity in the quest to obtain a required opinion from them, thus respondents names were not included in the questionnaire. Also, respondents were asked to freely quit from answering the questionnaire at their own discretion. In the final analysis, out of 400 sample respondents, 305 (representing 76%) responses were valid for data processing/ analysis. The authors then applied PLS-SEM for rigorous analysis while the software; ADANCO 2.O.1 was used and a bootstrapping resampling of 999 attempts. However, a quick preview of our field report shows that the majority of the respondents prefer local brand (64%) of fast-food as compared to the foreign brands (35%) available in the Ghanaian fast-food industry. Whilst the male category is the most dominant consumer of fast-food among the sampled population which, is quite not different from respondents who are unmarried (single). The summary of sociodemographics for the study respondents with respect to their frequency and percentage is given in Table 1 for easy understanding.

Data analysis technique
The test of the research model relied on Partial least square and structural equation modeling (PLS-SEM) using ADANCO Version 2.1. The authors used PLS-SEM over co-variance-based structural equation modeling (CB-SEM) because CB-SEM requires data to be normally distributed, while PLS-SEM holds no assumption about distributions of data. Thus, the overall results of a statistical test are not contradicted by non-normal data (Goodhue et al., 2012), hence the use of PLS-SEM. Regarding data analysis, we adopted the partial least squares (PLS) approach because it focuses on the explained variance of the criterion variable (J. F. Hair et al., 2019;J. Hair et al., 2017). Since this study is exploratory, the use of PLS is considered most appropriate by several marketing investigators among them (Amegbe & Osakwe, 2018;Chen et al., 2019;Henseler et al., 2016;Jibril et al., 2019). In this study, SmartPLS 3.2.9 was used for testing research hypotheses.

Measures and analytical approach
The items used for measuring the constructs were adapted from the existing literature. Measurements used in this study were adapted from previous studies with minor changes, these changes were particularly done with respect to the scales used in measuring the constructs. The new scales were adapted mainly to suit the current theme under study. It is therefore important to note that all the items were measured on a five-point Likert scale (1 = completely disagree, 2 = disagree, 3 = neutral, 4 = agree and 5 = completely agree). The five-point Likert scale was used mainly to determine respondents' level of opinions regarding how they agree or disagree with the measurement constructs. Also, all of the measurement items were in a positive statement. Therefore, Table 2, shows a summary of construct indicators questionnaire items with their respective literature sourced and a measurement scale.

Test of common method bias (CMB)
Taking inspiration from Podsakoff (2003), the authors stated in the present survey questionnaire particularly in the header section that there are no right or wrong answers to the questions asked. We further assured our respondents of their anonymity and also informed them that they were free to quit filling the questionnaire at any time. Moreover, as earlier stated, different anchors were employed in the questionnaire. Besides the qualitative measures taken to address potential concerns about CMB, this study, following the recent suggestions in the PLS-SEM literature and particularly Kock and Hadaya (2018), we employ the full collinearity approach, specifically variance inflation factor (VIF) for detecting evidence on CMB. The results of this post-hoc measure indicate that CMB is not a key concern since the computed VIFs are less than three (3) considering the maximum threshold of ten (10). Again, following previous research , the current work concludes that since a study that examines the presence of mediation effect, it is extremely difficult for respondents themselves to mentally manipulate. Therefore, the concerns about CMB are minimal here; hence, in this analysis, the potential for CMB is low.  (Kivela, 1997;Neumark-Sztainer et al., 1999) 6. Desire to select DES1: I enjoy eating at a famous fast-food joint. DES2: I'm willing to buy food from the fast-food joint. DES3: It is my pleasure to buy from a fast-food joint. (Ferguson et al., 2012;Kivela, 1997) 7. Advertising medium: TV & Radio Word-of-mouth RADTV1: TV and Radio adverts of a particular fast-food joint makes me go for that food. RADTV2: I enjoy watching TV cooking program and listening to LPM about the fast-food selection. WOM1: I follow a recommendation from friends when I want to buy from a fast-food joint. WOM2: My choice of a fast-food joint is possible when I hear people talking good about that food joint. (Auty, 1992;Mensah et al., 2017) Adzovie & Jibril, Cogent Social Sciences (2020)

Model assessment
According to pioneer scholars (see Hair et al., 2017), it is imperative to take inspiration from the PLS-SEM literature, the constructs' reliabilities were assessed thoroughly using Dijkstra-Henseler's rho along with Cronbach's alpha coefficients. As indicated in Table 3, all the values exceeded the threshold of 0.5 indicating strong coefficients of construct's reliability as suggested by Bagozzi and Yi (1988); Hair et al. (2019). The software ADANCO 2.0 version (Dijkstra & Henseler, 2015a, 2015b was used to evaluate the psychometric properties of the constructs and their underlying items. Using the Jöreskog's rho (pc) with a threshold of 0.7 and Dijkstra-Henseler's rho (ρA) with a threshold of 0.8 as a composite reliability determinant, our analysis fulfils those requirements. Therefore, composite reliability (CR) of the study constructs is presented by Dijkstra-Henseler's rho (ρA) with a minimum reliability coefficient 0.8883 and a maximum of 0.9301, while convergent validity, was presented by average variance extracted (AVE) which also exceeded the minimum threshold of 0.5 (see Table 3).
Again, with regards to the indicator loadings of the latent constructs, all items (indicator variables) were loaded meaningfully to their corresponding construct. The measured indicators have minimum loadings (coefficients) approximately 0.5 and maximum loadings of approximately 0.9. According to Bagozzi and Yi (1988) a loading above a threshold of 0.6 is the best measurement of a latent variable under study; therefore, factor loadings which were all greater than 0.6 (see Table 4). These indicator-variables have minimum loaded of 0.661 and a maximum load of 0.940 and this indicates measure what they ought to measure in practice. Hence, the summary of all the research constructs as well as their measurement items are shown in Table 4 with their corresponding loadings (coefficients). More so, the authors proceeded with a test of goodness of fit and the variance inflation factor. The variance inflation factor (VIF) was also performed to measure the amount of multicollinearity in a set of multiple regression variables. Mathematically, the VIF for a regression model variable is equal to the ratio of the overall model variance to the variance of a model that includes only that single independent variable. If the VIF is equal to 1 there is no multicollinearity among factors, but if the VIF is greater than 1, the predictors may be moderately correlated. A VIF between 5 and 10 indicates a high correlation that may be problematic. Therefore, per our estimations, VIF of our variables were less than 3; hence, there is no issue of multicollinearity (see Table 4).
Furthermore, to establish the discriminant validity on the other hand, Fornell and Larcker (1981) and a recent recommendations in the literature and in particular the Heterotrait-Monotrait (HTMT) ratio of correlations approach was adopted to assess the presence of discriminant validity among the latent variables (Henseler et al., 2015). Findings from Fornell-Lacker's criterion indicated that constructs satisfy both basic and stringent assumptions and this, therefore, establishes  Table 5) indicate AVE's of the measured constructs and must be greater than 0.5. At the same time, each construct's AVE should be of higher value (coefficient) at both column and row position over other constructs so that discriminant validity could be established (Fornell & Larcker, 1981). Whereas HTMT at a maximum value of 0.85 is the most conservative criterion, as it achieves the lowest specificity rates of all the simulation conditions. This means that any value of HTMT index greater than the threshold hold of 0.85 (Henseler et al., 2015) confirming that discriminant validity has been established, but in our case, there was no such presence of discriminant.
Squared correlations; AVE in the diagonal (in bold). Source: Authors' processing from ADANCO 2.1 Version

Structural equation modelling-Hypothesis testing
Proceeding from the assessment of the model fit, the structural model (path analysis) of the hypothetical analysis is required. Noticeably, it is relevant to achieve this stage of the analysis since it identifies and establishes the causal-effect or relationships of the constructs of the underlying research aim. As a matter of fact, results reveal direct and indirect effects of motivational factors that trigger customer selections cue towards a fast-food joint.

Direct effect
Our survey shows that customers' taste and preference (TASPRE) as well as time (TIM) have a significant positive relationship with the desire to select a particular fast-food joint among Ghanaian consumers, thus the regression coefficients and t-values showing; TASPRE (β = 0.4126, t = 6.6535) and TIM (β = 0.2329, t = 3.2182). Whilst the direct relationship between the constructs convenience (CONV), menu price (MEN) and the desire to select fast-food joint shows a positive effect yet, insignificant per the computations of their t-values (see Table 6). Again, the direct relationship/ effect between the constructs 'radio and television adverts (RADTV) and CONV (β = 0.3241, t = 4.9935), and that of RADTV and TASPRE (β = 0.3629, t = 5.5638) were all significant (see Table 7).

Indirect effect
The structural model further shows an indirect effect of the study constructs with regard to the desire/intent to select a fast-food joint in the Ghanaian context. Per our analysis, the results show that word-of-mouth (WoM) significantly predict customers' desire to select a fast-food joint through a mediating variable; taste and preference (TASPRE) with an estimation of (β = 0.1557, t = 3.9689), at the same time radio and television adverts (RADTV) in the fast-food market in Ghana significantly predict desire to select fast-food joint (DES) via a mediating variable; convenience (CONV) with an estimation of (β = 0.1693, t = 4.2977).

Coefficient of determination (R)
With regard to the predictive power of the study constructs, the coefficient of determination(R 2 ) of the regression model was assessed. The coefficient indicates the percentage of variation in the dependent variable that has been explained by the predictor (independent) variable. On the other hand, the Adjusted R 2 shows the amount of variance in the endogenous construct explained by the exogenous constructs. From Table 7, the estimated R 2 of CONV (0.105) showed a 10.5% of the variation in the CONV is explained by the construct RADTV (as an independent). Again, the R 2 of TASPRE (0.345) showed a 34.5% variation in the construct TASPRE is explained by the construct's RADTV and WoM. Whilst model R 2 of the dependent variable DES (0.460) indicates 46% of the total variation of the construct DES explained by the combined effect of individual constructs; CONV, TIMD, MEN, and TASPRE, and this could be traced in Table 7 and Figure 2.

Discussion and study implications
The aim of this study was to explore the motivational factors that actuate (or stimulate) consumers' intent to select a fast-food joint in an under-developed country setting, particularly, in Ghana, a sub-Sahara African region. At the same time, the partial goal of this survey was to examine the mediating role of two constructs; convenience (CONV), and taste and preference (TASPRE) among the underlying research constructs. The results revealed that "time" spent to go out for a fast-food as an alternative meal for either lunch or dinner plays a significant role with regard to consumers' decision-making. This cue suggests that time is not wasted when people decide to buy or select a fast-food joint as an alternative for their breakfast/lunch/dinner. Again, the study indicates that, during the working hours of the day, labourers/employees/apprentices are not willing (interested) in going to their domestic homes for lunch, since it is more convenient and less time-consuming to alternate such meal with fast-food, which according to their (customers) experience, it increases productivity at workplace. The study, therefore, corroborates the work of (Kivela, 1997;Mensah et al., 2017).
However, per the valid responses obtained for the study, menu price (MEN) was a positive determinant of buyers' intention towards a fast-food joint selection, but it was insignificant relative to other research constructs of this study. Since most of the respondents fall within the category of middleclass/workers, they are not perturbed about the price of the menu amongst these fast-food joints. Therefore, with respect to a cross-section of Ghanaian fast-food consumers, menu prices seemed not to be a major motivating factor for them to select a particular fast-food joint of their choice, and this analogy further supported the work of Dunn et al. (2011). Moreover, the fast-food industry continues to be a lucrative business all over the world. Therefore, practitioners of these activities should consider both extrinsic and intrinsic motivational factors that could help sustain their business.
Again, Word-of-mouth marketing (WOMM), also called the word of mouth advertising has gained a momentum in today's business environment, in that, it is actively influenced or encouraged by organizations such as "seeding" a message in a network rewarding regular consumers to engage in WOM, employing WOM "agents" among others. It is quite obvious to note that WOM in contemporary business has shown that there are three generic avenues to manage WOM for the purpose of WOMM. Among these include 1) Build a strong WOM foundation (e.g., sufficient levels of satisfaction, trust and commitment), 2) Indirect WOMM management which implies that managers only have a moderate amount of control (e.g., controversial advertising, teaser campaigns, customer membership clubs), 3) Direct WOMM management, which has higher levels of control (e.g., paid WOM "agents", "friend get friend" schemes). Also, Pro-consumer WOM has been suggested as a counterweight to the commercially motivated word of mouth. To this, our study revealed that WoM has a significant indirect effect on the choice of the fast-food joint selection through a mediating role of consumer's taste and preference for fast-food. The findings of the study, therefore, support the research hypotheses formulated and offer a theoretical contribution by enhancing knowledge of the relationships between the variables investigated as well as the factors that could influence WOM reliability. Hence, the significance of the research construct; WOM is consistent with the earlier works of (Bigne et al., 2018;Brown et al., 2007;Mensah et al., 2017). Theoretically, and though limited in scope, the findings of our study enhance the understanding of factors that influence consumers' desire to select a fast-food joint in general, and particularly in an under-developed country setting (like Ghana). This exploratory study gives leverage to interested scholars to re-examine the research model in a different cultural context in order to ensure the reliability and validity of the research constructs. The study further reminds academics that even though the presence of the internet has taken the media space; however, the traditional medium of marketing and advertisement in under-developed country settings still remains relevant. Again, this empirical evidence would shed more light on the concept; consumer behaviour in the fast-food industry as well as the significant role of the traditional medium of marketing communication, notably, Word-of-mouth, television, and radio adverts.
Practically, the findings also provide fast-food retailers/vendors with strategic guidelines for enhancing their communication strategies and delivering greater profits in a highly competitive environment. Moreover, fast-food vendors and social marketers particularly in the developing countries should further take advantage of the research results which suggest that word-of-mouth has the strongest relationship with source trustworthiness with regards to consumers' intent to select a fast-food joint. This is because the study also has a competitive meaning or understanding to fast-food retailers, as the model could be applied to formulate appropriate plans that could reach a large number of potential consumers in the fast-food industry.

Conclusion and future research directions
In this study, we explored the motivational factors that actuate (or stimulate) consumers' intent to select a fast-food joint in an under-developed country setting, particularly, in Ghana, a sub-Sahara African region. Additionally, the partial goal of this survey was to examine the mediating role of two constructs; convenience (CONV), and taste and preference (TASPRE) among the underlying research constructs. Using a quantitative research approach, a structured survey questionnaire was used to intercept buyers of fast-food at vantage points in the Cape Coast metropolis in the Central region of Ghana. A non-randomized sampling technique, precisely, the convenience sampling, was adopted to select popular fast-food joints that aided the researchers to intercept customers/buyers at the time of data collection. Application of partial least square and structural equation modelling (PLS-SEM) results from 305 valid responses revealed that the mediation (indirect) analysis supported all the mediated-hypotheses (H6 and H7) while only two direct hypotheses (H1 and H3) was not supported per our rigorous process and analysis at a significant level of 5% or t-value >1.96.
However, this study is not without limitation. First, given the sample size in this study, it is relatively small as fast-food vendors are scattered all over Ghana. Second but not last, the study only considered the research constructs from the perspective of fast-food buyers while neglecting the responses from the fast-food vendors. Therefore, it would be interesting if future researchers could increase the sample size so as to augment the current study. Notwithstanding, a future study could compliment both perspectives from fast-food vendors and fast-food buyers in a single study. Finally, future research could also test whether other retailers in the food industry could apply the same model to communicate with their market segments.