A hazard-based approach to modelling the effects of online shopping on intershopping duration

Previous work

At a conceptual level, transport researchers classified potential implications of ICT on personal travel into four types: substitution, complementarity, modification, and neutrality (Salomon 1985, 1986; Mokhtarian 2004; Hjorthol 2002; Mokhtarian 2009). Building on this categorisation, many researchers have attempted to quantify the net effects of increasing use of ICT on journey frequency and miles travelled. Bhat et al. (2003); Rotem-Mindali and Weltevreden (2013); Rotem-Mindali (2014) provided detailed reviews of such studies that focus more specifically on online shopping. Findings from this line of work, however, are diverse and sometimes contradictory. Results are heavily influenced by how net effects are being measured, what is being categorised as shopping (e.g., searching, transactions, returns), types of products being considered (e.g., furniture, electronics, books, groceries), sample selection, and other assumptions. Another related strand of transport literature investigates the relationship between frequencies of online shopping and traditional in-store shopping to gain further insights to the substitution or complementarity question (Cao 2012; Farag et al. 2007; Circella and Mokhtarian 2010; Zhou and Wang 2014). These studies also aim to explore how certain characteristics of individuals (e.g., socio-demographics, shopping attitudes, internet experience, geographical accessibility) will influence shopping frequencies via online and in-store channels using descriptive analyses, regression methods, and structural equations models. Most studies identify a positive relationship between online and in-store shopping frequency. For instance, Circella and Mokhtarian (2010) reported complementarity and found limited evidence for substitution between online and in-store shopping frequency for experience (clothing/shoes) and search (books/DVDs) goods. Farag (2006) found evidence of frequent in-store shoppers also shopping online more frequently, yet note that the causality is not easy to establish. More recently, Suel et al. (2015) reported evidence of substitution effects in the context of grocery shopping in London.

A common challenge is finding suitable datasets as national and regional travel surveys often capture very limited information regarding online activity and are subject to substantial local variation and idiosyncrasy. Circella and Mokhtarian (2010) and Cao (2012), for instance, relied on one-off ad-hoc survey data where respondents are asked to report how often they are engaged in online and in-store shopping separately on a frequency scale (e.g., ranging from ‘never’ to ‘more than once a week’). In the dataset used by Farag et al. (2007), online shopping frequency is captured using a points scale while information on in-store shopping is derived from a detailed two-day travel diary. Becky and Wang (2017) collected data on time spent for online shopping at home on a scale ranging from never to more than one hour per day. Zhou and Wang (2014) made use of the US National Household Travel Survey (NHTS) in this context, which benefits from a large sample size and detailed travel diary data for in-store shopping, yet only captures online shopping on a points scale. In short, none of the datasets used in cited studies contain information on actual online shopping activity, rather they make use of survey responses that measure online shopping activity levels on a points scale. Notwithstanding the limitations, these studies offered important insights to the complex relationships between online and in-store shopping. Importantly, they all highlighted the difficulty in finding a simple answer to the complementarity or substitution question, the value of recognizing the complexity of the underlying behaviour, and the need for tools to better understand changes triggered by wider use of technology.

Aim and specific objectives

Against this background, the overall aim of the present study is to investigate the effects of online shopping on the temporal patterns of shopping activity and related travel demand by making use of hazard models. Hazard based models have been previously applied to analyse inter-shopping duration for traditional in-store shopping trips (Kim and Park 1997; Leszczyc et al. 2000; Schonfelder and Axhausen 2001; Bhat et al. 2004b, 2005). In transport research, these studies were motivated by the need to develop multi-day (as opposed to single-day) activity generation model systems to account for weekly shopping patterns (Bhat et al. 2004a). The base line hazard function is used to model the increasing likelihood of participation to shopping with increase in elapsed time since the last shopping episode due to inventory depletion effects. However, to the best of our knowledge, previous work on inter-shopping duration modelling have not considered in-home shopping activity via online channels. Yet, one would expect participation to online shopping to also influence the likelihood of participation to in-store shopping and vice versa. Online shopping might also alter overall shopping patterns, leading to observed shifts in frequencies. One relevant study by Bhat et al. (2003) used hazard based models to analyse the effects of using mobile phones and internet on inter-shopping duration for non-maintenance shopping (excluding grocery and health related shopping etc.). Findings suggested potential substitution between mobile phone use and non-maintenance shopping travel. ICT-use data, however, was at the individual level relating to whether someone uses mobile phone and internet at all. Hence the effects of specific episodes of ICT use were not accounted for. Hazard-based models of inter-shopping duration can be extended to incorporate internet shopping where observed online shopping episodes influence the baseline hazard or modelled as end of duration events.

Shopping itself is a heterogeneous class of activities (Mokhtarian 2004; Visser and Lanzendorf 2004; Rotem-Mindali and Salomon 2007; Girard et al. 2003). For instance, individual behaviours differ substantially when shopping for daily groceries as compared to white goods purchases. Changes in retail supply, therefore, will also likely alter shopping behaviour in different ways in different sectors (e.g., electronics, clothing, furniture). While it is of interest in principle to investigate these differences, a comprehensive treatment is beyond the scope of a single paper. We limit our focus to grocery shopping occasions where a transaction occurs (i.e., excluding pre- and post-purchase trips). The reasons for this decision is as follows. First, it is the most common and frequent type of shopping, hence more relevant for travel implications¹ (Golob 2002). Second, online grocery market is a fast growing segment at present and is expected to grow with an annual growth rate of more than 50% over the next five years in the UK (Gladding 2016). Hence, potential implications on personal travel and logistics operations will be increasingly important. Third, since grocery shopping has minimal recreational value it allows to really get at the physical vs. virtual issue net of any of the complicating factors relating to recreational value of shopping (Gould and Golob 1997). Fourth, pre-purchase (searching) and post-purchase (returns) stages are less relevant for additional trip generation reducing the complexity involved in modelling and data collection. Fifth, logistics operations for grocery shopping deliveries are different in nature and often more cumbersome (e.g., tighter delivery windows, constraint for in-person deliveries, fresh products). Therefore, policy implications of online grocery shopping associated with extra freight vehicle trips will not only be important but also different when compared to other types of products.

More specifically, the objectives in the present are as follows. First is to incorporate online shopping activity in hazard-based inter-shopping duration models for grocery shopping. This requires a data source where individual shopping events across both online and in-store channels are captured. Second objective is to analyse effects of online shopping separately on (i) inter-shopping duration for all types of shopping events (including both online and in-store activity) and (ii) duration between in-store shopping trips (i.e., out-of-home shopping activity) generating personal travel to stores.

Modelling the duration between consecutive shopping episodes including both online and in-store activity will contribute to multi-day activity generation work for estimating travel demand. It will also help explore the effects of online shopping on frequency and trips generated, potentially providing new insights to the complementarity or substitution question. It will also contribute to the development of tools that enable predicting when the next shopping event or trip will take place. Such prediction capability is crucial for retail delivery operations especially in certain market segments such as online groceries given their struggle to make businesses profitable (Twentyman 2015).

Data

The UK has a comparatively developed online market for groceries (Kantar Worldpanel 2015), hence offers an opportunity to study impacts of ICT use on personal travel patterns for the purposes of the present study. National and regional travel surveys (e.g., British National Travel Survey (DfT 2016), London Travel Demand Survey (TfL 2016)) often used in transport research offer detailed data on physical trips for food shopping, yet capture very limited information regarding online activity. Specific episodes are not recorded and information regarding when each online shopping event took place, crucial for our purposes here, is not available.

Data used for the present study were obtained from a consumer panel run by Kantar Worldpanel (KWP). Participating households use hand-held optical scanners to record daily purchases of fast moving consumer goods that are brought home. Fast moving consumer goods include products found in supermarkets typically bought frequently and at relatively low prices. Examples are groceries, toiletries, health and beauty items. Panellists are also asked to scan and send till receipts whenever they make a shopping trip or get deliveries. Shopping occasions are recorded regardless of the retail chain, hence include visits to smaller local shops, independent corner shops, and larger supermarket chains. KWP is a continuous panel where households can participate for as long as they wish providing longitudinal data and participants receive incentives in the form of redeemable points. Household characteristics including socio-demographics are collected at the initial interview and continuously updated every six-months or annually. All household members record their purchases separately, however the dataset employed had an indicator for main shopper only but otherwise did not distinguish between other household members. KWP defines main shoppers as household members that are responsible for the bulk of grocery shopping in their household. For model estimations, we use shopping occasions recorded by all shoppers in the household without distinguishing between different members of the household. Purchases from all retailers including online retailers are recorded together with basket sizes in terms of monetary value. Unlike travel surveys, both online and in-store shopping observations contain data at the same level of detail. Shopping records include visits where a transaction occurs since product level purchases are recorded, hence will exclude search only or returns only visits. Since the focus is on grocery shopping, however, shopping occasions which will not involve any purchases are likely minimal. Information on basket sizes is also crucial as the amount of shopping will also potentially influence the duration until the next shopping occasion. For assessing the influence of online shopping on inter-shopping duration, it is important to separately account for basket size effects especially since online shopping is often associated with large basket sizes due to minimum size requirements for delivery. Consumer panel data is also attractive as market research companies operate consumer panels in many countries and some make it available to researchers.² Availability is important for generalisability of our proposed method. Consumer panels, however, were also found to have some shortcomings. Information on the retail chain is available, but the specific store is not known. They also do not gather travel related information, for instance, travel mode choice for shopping is not collected that is potentially important when analysing travel implications. Market research companies collect relatively limited demographics information on respondents and transitions (e.g., changes in employment, income, marital status, children) are not always well recorded.

The sample obtained from KWP covers a one year period between September 2013 and August 2014 from a sample of 168 households in London. 124 households in our analysis sample are panellists who reside in the two selected boroughs, Barnet and Enfield. Thirty four out of 124 were online shopping households with reported online observations during the one year period. An additional forty four households were drawn randomly from London among online shopping households since we are primarily interested in their behaviour. This, however, causes sampling bias as online shopping households are over represented in our estimation sample. To correct for this, we use weights for all observations assuming that the share of online shopping households in London matches observed shares in Enfield and Barnet.³ When we compare demographics for our sample and Census data from Enfield and Barnet in 2011 (Office for National Statistics 2011), we found that households with older main shoppers and highest social classes are overrepresented in our study sample (Suel 2016). We note that no information is available from the Census on main shoppers, hence age of household reference person is used as proxy for comparison.

While non-traditional panel data offers significant advantages for research, potential problems and sampling biases should not be overlooked. We note that our sample will share important characteristics relating to the geographical proximity of residential addresses. Also, while the number of recurring observations from each responding household was large, the main limitation is that the number of households in the sample was relatively small due to budget constraints. If population scale prediction is a prioriry, then a larger and more representative sample would be desirable. Here, we demonstrate the insights drawn and prediction capabilities that can be developed with existing data sources using modelling tools as presented below.

Analysis methods

Hazard-based methods have been initially developed for modelling duration data such as time to failure or time to some form of state change. These methods are often used in survival analysis in the context of biological problems (e.g., expected duration of time to death or organ failure) and reliability analysis for mechanical systems. Hensher and Mannering (1994); Bhat and Pinjari (2000) presented extensive reviews of hazard-based duration models and their applications in transport research. The basic idea is to model the probability of an end-of-duration occurrence given that the duration lasted for some time. This probability will depend on the length of elapsed time and also on relevant co-variates. These models are suitable for analysing inter-shopping duration, i.e. time between two consecutive shopping occasions. The probability of ending a duration since last shopping activity is dependent on length of the duration (time elapsed since last activity) due to depletion effects (i.e. an individual is more likely to go grocery shopping on any given day if s/he has not done so for a longer period). This duration will also likely depend on other observable characteristics such as household size (larger households consume more and might need to go shopping more frequently) and basket size on last activity (if more products are stocked the need to go shopping again might decrease).

For our empirical analyses, we differentiate between shopping events and shopping trips. For the former, gap times are defined as time between two shopping events regardless of whether they were online or in-store. Online and in-store observations are both modelled as end-of-duration occurrences. For the latter, on the other hand, gap times are defined as time elapsed between two in-store shopping trips. Only in-store observations, which generate personal trips to physical stores, are modelled as end-of-duration occurrences. From available data, we calculated times between shopping events and shopping trips separately over one year. These are also used as durations for hazard based analyses. Figure 1 shows the distribution of inter-shopping-event and inter-shopping-trip times over our sample. As expected, times between shopping trips are longer in duration on average and there are cases where households do only online shopping for long periods without physically visiting stores. Note that there were also more outlier cases for inter-shipping-trip durations grouped in 14+ days bin with more cases where time between consecutive shopping trip exceed two weeks. When multiple shopping events or trips are observed in a single day, durations are computed as being evenly distributed across one day. The first bins in histogram plots include shopping activity observed within one day, which makes up almost 35% of all observations for each category.

Open image in new window

Our main interest in this paper is to test the effect of online shopping activity on the instantaneous rate of occurrence of shopping (inter-shopping-event durations) and travel for shopping (inter-shopping-trip durations). We do so by incorporating, along with other shopper characteristics and situational variables, two key indicators: (i) whether the household have adopted online shopping as a channel; and (ii) whether their last shopping was online. The former is a dummy variable and is equal to one only after a respondent is observed shopping online and remains equal to one for the remainder of our observation period.⁴ The former variable therefore captures a general long term effect of adoption, whereas the latter would reveal short term (tactical) effects of buying online.