본 연구는 옵션의 정렬가능성과 함께 옵션 프레이밍 유형이 기본제품과 증대제품의 평가에 미치는 차별적 영향을 두 가지 상황-즉, 제품에 대한 평가 점수가 추가 정보로 제공되지 않는 상황과 추가 정보로 제공되는 상황- 으로 나누어서 살펴보았다. 네 차례의 실험 결과, 첫째, 제품 평가 점수가 추가 정보로 제공되지 않는 상황에서 옵션이 정렬 가능한 경우에 기본제품이 독립적으로 제시된 조건의 소비자는 상향 옵션 프레이밍 조건과 하향 옵션 프레이밍 조건의 소비자보다 기본제품을 높게 평가하는 것으로 나타났고, 상향 옵션 프레이밍 조건의 소비자 는 하향 옵션 프레이밍 조건의 소비자보다 기본제품을 높게 평가하는 것으로 나타났다. 이에 반해서 증대제품의 평가에서는 상향 옵션 프레이밍 조건과 하향 옵션 프레이밍 조건의 소비자가 증대제품이 독립적으로 제시된 조 건의 소비자보다 한계적으로 유의한 수준에서 증대제품을 높게 평가하는 것으로 나타났다. 둘째, 제품 평가 점 수가 추가 정보로 제공되는 상황에서 옵션이 정렬 가능한 경우에 기본제품이 독립적으로 제시된 조건의 소비자 와 상향 옵션 프레이밍 조건의 소비자는 하향 옵션 프레이밍 조건의 소비자보다 기본제품을 높게 평가하는 것으로 나타난 반면, 증대제품의 평가에서는 증대제품이 독립적으로 제시된 조건의 소비자, 상향 옵션 프레이밍 조 건의 소비자, 그리고 하향 옵션 프레이밍 조건의 소비자 간에 유의한 차이가 없는 것으로 확인되었다. 셋째, 제 품 평가 점수가 추가 정보로 제공되지 않는 상황에서 옵션이 정렬 불가능한 경우에 상향 옵션 프레이밍 조건의 소비자는 기본제품이 독립적으로 제시된 조건의 소비자와 하향 옵션 프레이밍 조건의 소비자보다 기본제품을 높게 평가하는 것으로 확인되었고, 기본제품이 독립적으로 제시된 조건의 소비자는 하향 옵션 프레이밍 조건의 소비자보다 기본제품을 높게 평가하는 방향성은 나타났지만 통계적으로 유의하지는 않았다. 또한 증대제품의 평가에서 상향 옵션 프레이밍 조건의 소비자는 증대제품이 독립적으로 제시된 조건의 소비자보다 한계적으로 유의한 수준에서 증대제품을 높게 평가하는 것으로 나타났고, 상향 옵션 프레이밍 조건의 소비자는 하향 옵션 프레이밍 조건의 소비자보다 증대제품을 높게 평가하는 것으로 나타났다. 넷째, 제품 평가 점수가 추가 정보로 제공되는 상황에서 옵션이 정렬 불가능한 경우에 상향 옵션 프레이밍 조건의 소비자는 기본제품이 독립적으로 제시된 조건의 소비자와 하향 옵션 프레이밍 조건의 소비자보다 기본제품을 높게 평가하는 것으로 나타났고, 기본제품이 독립적으로 제시된 조건의 소비자는 하향 옵션 프레이밍 조건의 소비자보다 기본제품을 높게 평가하는 것으로 확인되었다. 그리고 증대제품의 평가에서는 상향 옵션 프레이밍 조건의 소비자가 증대제품이 독립적으로 제시된 조건의 소비자보다 한계적으로 유의한 수준에서 증대제품을 높게 평가하는 것으로 나타났으며, 상 향 옵션 프레이밍 조건의 소비자가 하향 옵션 프레이밍 조건의 소비자보다 증대제품을 높게 평가하는 것으로 확 인되었다. 마지막으로 본 연구결과가 제시하는 이론적?관리적 시사점들을 제안하고, 연구의 한계와 향후 연구 방향에 대해서도 논의하였다.
Add-on features are discretionary benefits that provide utility only if consumed with the corresponding base good(Guiltinam 1987). For example, manufacturers of electronic equipment such as digital cameras and laptop computers produce or source a wide range of accessories, including carrying cases and memory cards. Previous research in marketing has shown that consumers often draw inferences about a product`s utility on the basis of contextual cues (e.g., add-on features) (Bettman, Luce, and Payne 1998; Broniarczyk and Alba 1994; Huber and McCann 1982). In Bertini, Ofek, and Ariely`s (2009) research, the experimental results lend further support to this idea and make a compelling case that add-on features can influence consumer behavior beyond what their inherent value would suggest. Bertini et al.(2009) classified add-ons according to the type of improvement they provide: alignable add-ons enhance existing product features, and nonalignable add-ons introduce new features. The distinction between alignable and nonalignable add-ons is important because consumers are likely to use different cognitive processes to assess these two types of improvements. A substantial number of studies show that evaluation shifts from dimensional to holistic processing as the alignability of objects decreases (e.g., Johnson 1984; Payne 1982; Russo and Dosher 1983; Slovic and MacPhillamy 1974). Building on this basic finding, Bertini and colleagues predicted that alignable add-ons would affect evaluation by shifting the reference level of the same attributes they modify (an attribute-level inference), whereas nonalignable add-ons would have an impact by cueing more general, attitude-based inferences about product value. Specifically, Bertini et al.(2009) believe the presence of an alignable add-on establishes a range of attribute values that consumers then use to judge the performance of the base good. The endpoints of this range are the level initially specified in the product (the lower bound) and the level obtainable by purchasing the add-on (the upper bound). Range theory posits that the attractiveness of any stimulus is inferred by its position within the range of possible values (e.g., Yeung and Soman 2005), which implies that a product`s original attribute level will be judged less favorably when an alignable add-on is available than when no such option exists. Conversely, the inferential process underlying the case of nonalignable add-ons is expected to be different. Although consumers now lack a natural frame of reference with which to judge this improvement, its novelty is likely to make cognitions about the add-on salient at the time of evaluation. General attitude toward a salient object have been found to trigger similar attitudes toward broader, related objects (e.g., Holbrook 1983)-a result commonly referred to as the "halo effect." Since in most situations nonalignable add-ons are viewed favorably by consumers, Bertini et al.(2009) proposed that a consumer`s positive attitudes toward a nonalignable add-on can consequently shape positive attitudes toward the base good. In sum, it was found that offering alignable add-ons can render a base good less appealing, whereas offering nonalignable add-ons has the exact opposite effect. Moreover, prior research has found that contextual inferences in decision making occur predominantly when people lack sufficient knowledge to assess alternatives with confidence (e.g., Bettman et al. 1998; Broniarczyk and Alba 1994). Thus, Bertini et al.(2009) proposed that the predicted effects of add-ons on evaluation should be contingent on the amount of product information available to consumers in the marketplace and that additional information reduces uncertainty and therefore weakens or even cancels out any potential impact. That is, it was found that, irrespective of type, add-ons have no impact on evaluation when people possess sufficient information to judge a product`s value independently. In addition to the alignability of options (features), however, the current research focuses on the differential effects of option framing on consumer evaluations of the base product and the augmented product (the base product plus the extra features). Generally, consumers are allowed to create their own consumable products by either starting with a basic product and adding desired features (i.e., upgrade option framing) or starting with a fully loaded product and deleting undesired features (i.e., downgrade option framing; e.g., Levin, Schreiber, Lauriola, and Gaeth 2002; Park, Jun, and MacInnis 2000). Much prior research concerning the differential effects of option framing (upgrade vs. downgrade) on choice-that is, consumers would tend to choose more options if they were to use downgrade as opposed to upgrade option framing has been explained by many different accounts. In particular, loss aversion or an endowment-based account (cf. Kahneman, Knetch, and Thaler 1991) clearly predicts that downgrade option framing would have a stronger effect on choice than upgrade option framing (e.g., Levin et al. 2002; Park et al. 2000). Taken together, therefore, this research examines the differential effects of option alignability (alignable vs. nonalignable) and types of option framing on consumer evaluations of the base product and the augmented product when additional product information (i.e., product`s quality rating) is available or not. A base product (hereafter termed a control 1 condition) or an augmented product (hereafter termed a control 2 condition) can be presented independently or in different option framing conditions (i.e., upgrade or downgrade). In the current research, our predictions are as follows: First, when options are alignable, a base product will be evaluated more positively in the control 1 condition than in both the upgrade option framing condition (hypothesis 1-1) and the downgrade option framing condition (hypothesis 1-2). When options are alignable, a base product will be evaluated more positively in the upgrade option framing condition than in the downgrade option framing condition (hypothesis 1-3). Second, when options are alignable, an augmented product will be evaluated more positively in the upgrade option framing condition than in the control 2 condition (hypothesis 2-1). When options are alignable, an augmented product will be evaluated more positively in the downgrade option framing condition than in the control 2 condition (hypothesis 2-2). Third, when options are nonalignable, a base product will be evaluated more positively in the upgrade option framing condition than in the control 1 condition (hypothesis 3-1). When options are nonalignable, a base product will be evaluated more positively in the upgrade option framing condition than in the downgrade option framing condition (hypothesis 3-2). When options are nonalignable, a base product will be evaluated more positively in the control 1 condition than in the downgrade option framing condition (hypothesis 3-3). Fourth, when options are nonalignable, an augmented product will be evaluated more positively in the upgrade option framing condition than in the control 2 condition (hypothesis 4-1). When options are nonalignable, an augmented product will be evaluated more positively in the upgrade option framing condition than in the downgrade option framing condition (hypothesis 4-2). Fifth, when additional product information about a product`s quality rating is available and options are alignable, a base product will be evaluated more positively in the control 1 condition than in the downgrade option framing condition (hypothesis 5-1). When additional product information about a product`s quality rating is available and options are alignable, a base product will be evaluated more positively in the upgrade option framing condition than in the downgrade option framing condition (hypothesis 5-2). Sixth, when additional product information about a product`s quality rating is available and options are alignable, there will be no difference in evaluations of an augmented product among the three conditions (i.e., upgrade option framing condition, downgrade option framing condition, and control 2 condition) (hypothesis 6). Seventh, when additional product information about a product`s quality rating is available and options are nonalignable, a base product will be evaluated more positively in the upgrade option framing condition than in the control 1 condition (hypothesis 7-1). When additional product information about a product`s quality rating is available and options are nonalignable, a base product will be evaluated more positively in the upgrade option framing condition than in the downgrade option framing condition (hypothesis 7-2). When additional product information about a product`s quality rating is available and options are nonalignable, a base product will be evaluated more positively in the control 1 condition than in the downgrade option framing condition (hypothesis 7-3). Finally, when additional product information about a product`s quality rating is available and options are nonalignable, an augmented product will be evaluated more positively in the upgrade option framing condition than in the control 2 condition (hypothesis 8-1). When additional product information about a product`s quality rating is available and options are nonalignable, an augmented product will be evaluated more positively in the upgrade option framing condition than in the downgrade option framing condition (hypothesis 8-2). In this research, four experiments were conducted to test and confirm these predictions. Finally, we discuss the theoretical and practical implications of our findings and develop directions for future research.