Abstract
Shopping is one of the most important parts in our daily life. We, as consumers, shop almost every day, whether it is a quick stop to the seven-eleven down the corner, or a family shopping in a large store like Wal-Mart at the weekend. From our perspective, when we plan to shop, all we want to consider is what we are getting and how much those items cost. On the other hand, as the suppliers in the shopping chain, the owners or investors of the stores have much more to consider when, where or even which kind of store they run or invest. The data also provides information for stores to consider the public transport system to attract more consumers. For this empirical project, our group decided to focus on the relationships between household’s grocery spending and some factors of that person who takes this survey. To be specific, we concentrated on three factors. First is the consumer’s gender. Second if he/she owns a car. Finally, how his/her trip time from home to the store related to the total spending. We used the multiple regression models to include these factors as different regressor of the grocery spending.
So far, only hypothesis 1, the one about cars proved to be valid. The remaining two hypotheses about gender and distance and trip time were all invalidated. Key findings include: people who own cars end up spending more on household groceries than those who do not. Females tend to spend less on household groceries compared to males. And the farther the shopper is from the store location, the less he or she tends to spend on the actual household grocery shopping.
Introduction
This paper focuses on the two groups of variables namely shopping spending, particularly on grocery and other related groups; the second group of variables would be factors that may be used to describe and or stratify the participants of this survey. To specify the factors listed under the second group of variables, they are gender, car ownership, and distance to the nearest grocery store or any shopping chain. The objective of this study is to describe the type of relationship that exists between these two groups of variables. For starters, the author of the paper already assumes that there exists a correlation between these two variable sets. The significance of the study is centered on describing what kind of relationship that is. The paper was divided into various sections namely the introduction, econometric model (including a discussion of the chosen dependent and independent variables), data analysis, results and discussions, and conclusion. The scope of this paper can be fairly limited by the chosen factors and variables. This means that the author only focused on grocery shopping. There are, for example, numerous forms of shopping, especially in this modern age and time. There are online shopping, window shopping, and retail shopping. Grocery shopping is a more narrowed down version of these two. Additionally, these different types of shopping practices can be differentiated by the factors that motivate the consumers. Consumers who engage in retail shopping, for example, do so primarily because they see the habit of shopping as a way to relieve stress and experience a certain feeling of satisfaction; those who engage in grocery shopping, on the other hand, do so out of necessity. Grocery items include food, hygiene products, and other commodities that a typical household cannot do without. Therefore, at some point, they will eventually find themselves in one of the nearby grocery stores. Although spending and shopping motivation can significantly impact an individual’s buying decisions, it does not tell the entire story. In the case of this study, for example, the focus is not on what motivates the grocery shoppers to spend money on items but what factors affect their level of spending. This means that the essential goal of this study is to describe an already recognized relationship between two set of variables.
Research Question
Wealth equality is one of the most significant issues in society today. How much an individual or an entire household spends for his grocery can be used as an indicator of wealth and or financial literacy. A household belonging to the high income-earning bracket that only spends a small percentage of its total disposable income for groceries even though the nominal value of such expenses are high compared to another household belonging to the low income-earning bracket may still be considered financially sound and literate. Another household who spends roughly the same (in terms of nominal value) for grocery shopping but belongs to a lower income-earning bracket may not be considered financially sound and literate. This is just one of the many ways how the findings from this study can be used by other researchers. It would not only be interesting but certainly helpful especially for other researchers to know what factors and variables affect the level of spending of certain individuals for grocery shopping. The research question that this study aimed to answer was the one that asks what kind of relationship exist between household grocery spending level and car ownership, gender, and distance of the consumer from the nearest grocery store.
Review of Related Literatures
Numerous literatures exist that may explain how each of the variables in the second group may affect household grocery spending but the consensus is that they are correlated in some way. The fact that numerous studies that link any two of these household grocery spending-related variables together may already be a direct evidence of that. In a study published in the California Management Review, for example, the researchers examined the different factors that affect store choice and shopping behavior within the context of various price formats. They argued that different consumers are attracted to different price formats and that certain people, based on a host of factors including but not limited to what type of shopper they are and their socioeconomic status, are more likely to subscribe to certain price formats. People who shop for pleasure and belong to the upper socioeconomic classes, for example, are more likely to subscribe to low price formats, among other observations . Relating this evidence to this study, car ownership may be seen as a representation of socioeconomic status in that persons who own a car may be identified as those belonging to higher socioeconomic classes while those who do not own one may be linked to the lower classes. If the study published in the California Management Review is to be used as a basis, then it would follow that people who own cars may turn out to have a higher level of spending on household grocery shopping than those who do not.
In another study published in the International Journal of Research in Marketing, the researchers investigated the impact of the use of promotion-based store management and multiple-store shopping on the shopping and spending patterns and habits of the consumers. The researchers in that study argued that it is likely for modern day shoppers to engage in a shopping habit that makes use of single-purpose multiple-store shopping. This is where the shoppers only shop for one reason yet they end up visiting more than one shop. What they found out in their study was that “consumers may systematically visit multiple stores to take advantage of two types of store complementarily; in the case of fixed cost complementarity, consumers alternate visit to high and low fix cost stores to balance transportation and holding costs against acquisition costs” . This can be interpreted in various ways. One possible way to interpret this in relation to the conceptual framework of this paper is to draw upon the store travel distance as a factor—the distance that a grocery shopper has to cover before he arrives from his home to the grocery shopping floor. If this study from the International Journal of Research and Marketing is to serve as a basis, then it would be safe to suggest that although distance may affect the level of spending that a family may have for groceries, it is just one of the many factors that can affect it. For example, in the generation of the assumption, the author of this paper has assumed that the typical grocery shopper would only go to one store location to buy all of the grocery items he and his family members need. While this may be the case for some people, the study by Gijsbrechts, Campo, and Nisol contradicts this, especially if there are a lot of other grocery stores nearby that offer the same level of value—in which case, the customers may resort to multiple-store visits in an effort to squeeze out all of the valuable offers in the stores located near them in situations where they think single store visits may not be enough to maximize their shopping savings. Then again, in a single store visit scenario—where it can be guaranteed that the shoppers would only visit a single store and not a collection of it within the area, distance to the shopping center may indeed be a highly significant variable, one that consumers would carefully consider when they are computing their shopping expenses. According to a study that was published in the Journal of Marketing, in-store travel distance may directly affect shopping spending levels in a directly proportional manner. The type of spending that the researchers examined there was unplanned spending noting that the farther the store’s distance is from the shoppers’ location, the more likely that they would spend and shop impulsively, both of which leads to higher overall grocery spending levels .
And the last among the list of variables that the author of this paper is trying to relate to household grocery spending would be the gender of the person doing the shopping. This is, by far, the most complex variable to be studied among the three. This can be evidenced by the sheer abundance of references linking gender and spending habits and shopping orientations alone. In a study published in Sex Roles, for example, the researchers examined the impact of gender differences on attitudes towards the two main types of buying namely conventional and on-line shopping. Some of the most important findings in that study suggest that “the online environment has an effect on buying attitudes by more strongly so for women than for men; whereas men’s functional concerns are amplified rather than changed in the shift from conventional to on-line buying, women’s motivational priorities show a reversal, and less involvement in shopping” . This only shows that women have a higher likelihood to spend more when it comes to grocery shopping than men—as evidenced by their more pronounced reactions to marketing strategies compared to men. Men, based on what Dittmar, Long, and Meek’s study suggested, are more of a utilitarian buyer, sticking to their planned shopping budget, although it would be unsafe to generalize and stereotype based on gender alone.
Model
There are a total of three possible connections that can be made in this study and those connections would only be between any of these four variables: household grocery spending, car ownership, distance to the grocery shopping location, and gender of the person doing the shopping. The dependent variable would be the household grocery spending level; it would be dependent on the three independent variables namely car ownership, distance to the grocery shopping location, and the gender of the person doing the shopping.
Considering that there are numerous links and connections that are being examined, it would only be logical to have more than one research hypothesis in this study. To specify the accuracy of the hypotheses, one hypothesis was allocated for each of the independent variables—as their relationship with the dependent variable is examined. This way, it would be easier to specify what caused what in certain observations.
Higher levels of household grocery spending can be observed among individuals who own cars
Higher levels of household grocery spending can be observed among individuals who live farther from the grocery store locations
Higher levels of spending can be observed among female household grocery shoppers
It is worth noting that all of these specific hypotheses are based on the consensus generated during the review of related literatures. This means that should most, if not all, of these hypotheses turn out to be true, that would show how inline this study is with other previously-published studies. Ideally, that would be the case for this paper.
Results
Consumerism has created the type of society that people know of today—the one where everyone consumes products and commodities in order to survive. This phenomenon has made the act of going to and from various shopping locations to shop a ubiquitous activity. Considering this, grocery shopping is something that is done because it is a requirement and not just a privilege. Unfortunately, knowing that people, at some point in their life, would be a part of the cycle of consumerism is not enough. What is more important at this point is to study how certain factors affect their way of shopping, or in this case, their level of spending when it comes to grocery shopping. In this paper, the author examined three independent variables and their impact on the level of spending for groceries. These three variables were car ownership, distance from the store location, and gender of the person doing the shopping.
Car Ownership
The research hypothesis for car ownership, based on the review of related literatures suggests that higher levels of household grocery spending can be observed among individuals who own a car. Using the regression analysis method proposed earlier, what the author should expect to find in order to validate this research hypothesis is a negative coefficient for car. A negative coefficient for this variable—in relation to household grocery spending (that has a coefficient of 73.55), would mean that individuals who own a car indeed end up spending more for their household grocery compared to those who do not own one.
Below is a table that summarizes the results based on STATA—the tool that was used to do the computations.
As shown here, specifically C1, the coefficient was negative and this only means that the car ownership hypothesis is valid.
Distance and Trip Time
The research hypothesis dedicated for trip time suggests that a higher level of household grocery spending can be observed among individuals who live farther from the store location. The opposite of the principle used in validating the hypothesis for car ownership was applied here. What the author of the paper expects to find is a positive coefficient for distance and trip time because the appearance of which would mean that the spending levels for household grocery for those living farther (i.e. longer trip time) from the store locations end up spending more. That coefficient (i.e. trip time) turned out to be -0.031. This means that the research hypothesis for distance and trip time is invalid. A negative coefficient for trip time means that for every one more minute of trip time the shopper spends on the road, less money is being spent on groceries in order to offset the additional cost of time lost and for transportation. Specifically, $0.031 is being spent less on groceries and more on transportation every minute lost in the road for trip time.
Gender
The research hypothesis dedicated for gender suggests that a higher level of household grocery spending can be observed among females than males. The same principle used in validating the hypothesis for distance and trip time was used here; specifically, what the author expected to find was a positive coefficient because that would mean that the females would indeed have a higher level of grocery spending than males. In the analysis, however, the coefficient turned out to be negative at -0.27. A negative coefficient means that females are spending less on household grocery than their male counterparts. This only show that the research hypothesis on gender and its effects on household grocery spending that is based on previously published literatures is invalid.
Data Analysis
Shopping is one of the most important parts in our daily life. We, as consumers, shop almost every day, whether it is a quick stop to the seven-eleven down the corner, or a family shopping in a large store like Wal-Mart at the weekend. From our perspective, when we plan to shop, all we want to consider is what we are getting and how much those items cost. On the other hand, as the suppliers in the shopping chain, the owners or investors of the stores have much more to consider when, where or even which kind of store they run or invest. The data also provides information for stores to consider the public transport system to attract more consumers.
For this empirical project, our group decided to focus on the relationships between household’s grocery spending and some factors of that person who takes this survey. To be specific, we concentrated on three factors. First is the consumer’s gender. Second if he/she owns a car. Finally, how his/her trip time from home to the store related to the total spending. We use the multiple regression models to include these factors as different regressor of the grocery spending. In the survey, for gender and car ownership, the actual answers provided are not integers but literal words. In order to analyze these factors in STATA, binary variable or dummy variables must be applied. We define 1 if that person does not have a car, 0 otherwise; similarly, 1 if that person is female, 0 means he is male. Therefore we can transfer literal words into STATA and regression model.
With the help of STATA, we derived the population regression function as {grocery spend=73.55136-1.5385*car (=1 if no car, =0 if owns a car)-0.2695*gender (=1 if female, =0 if male)-0.0308*trip time}. 73.55136 can be considered as the constant regressor or the intercept of the expected value of grocery spending when all the factors equal to zero. We can see that the coefficient of car is negative which indicates that holding all other factors constant, having a car leads to more grocery spending, and more specifically, 1.5385 dollars more. When we drive to the store, the car cost gasoline and other cost which become a negative factor in grocery spending. The coefficient of the gender is negative, meaning that the females tend to have lower grocery spending than male if holding all other regressors constant. The regressor trip time stands for that person’s time from his/her home or workplace to the store, or in other word, how far the store is for a household. The coefficient -0.0308 means that for every one more minute or trip time to the store, less 0.0308 dollars is being spent on grocery. Another way to interpret this is holding all other regressors constant, the further the store is, the less people will spend on grocery. In reality, it totally makes sense because we usually go to the nearest store if we just want to pick up some grocery. When we travel longer time to a further store, it is usually for other purposes that local grocery stores cannot satisfy us.
Choosing the right regressor
In theory, the right regressor was chosen here—i.e. household grocery spending. This is because it was the dependent variable and so the variables being studied should all be studied in relation to it. This can be evidenced by the R-squared value, which is a number between 0 and 100%. The higher it is the better because it means that the data analysis model chosen fits the type of data being studied. In this case, the R-squared value was 0.02%. It still rests in the positive territory. Combining the R-squared value and the theory that the dependent variable should indeed be chosen as the regressor in the study, it would be safe to say that the right regressor was chosen.
Conclusions
Analyzing people’s shopping patter is extremely helpful not only for retail investment but also for the neighborhood community. So far, only hypothesis 1, the one about cars proved to be valid. The remaining two hypotheses about gender and distance and trip time were all invalidated. Key findings include: people who own cars end up spending more on household groceries than those who do not. Females tend to spend less on household groceries compared to males. And the farther the shopper is from the store location, the less he or she tends to spend on the actual household grocery shopping. For future research, a more appropriate approach would be to use a single set of samples only and a single pair of dependent and independent variable so that the researchers can focus on studying them and to avoid confusion.
Appendices
Car Ownership
Gender
Grocery Spending
Distance and Trip Time
Works Cited
Dittmar, Helga, Karen Long and Rosie Meek. "Buying on the Internet: Gender Differences in On-line and Conventional Buying Motivations." Sex Roles (2004): 423-444. Print. 05 June 16.
Gijsbrechts, Els, Katia Campo and Patricia Nisol. "Beyond promotion-based store switching: Antecedents and Patterns of Systematic Multiple Store Shopping." International Journal of Research in Marketing (2008): 05-21. Print. 05 June 16.
Hui, Sam, et al. "The Effect of In-store Travel Distance on Unplanned Spending: Applications to Mobile Promotion Strategies." Journal of Marketing (2013): 01-16. Print. 05 June 16.
Tang, Christopher, David Bell and Ho Teck. "Store Choice and Shopping Behavior: How Price Format Works." California Management Review (2001): 56-74. Print. 05 June 16.
