사전지식 및 관여도가 소비자의 신념불일치에 미치는 효과

한웅희(Woong Hee Han),김재일(Jae Il Kim),이유재(You Jae Yi),박기완(Ki Wan Park) 한국마케팅학회 2009 마케팅연구 Vol.24 No.1

본 연구에서는 어떤 대상에 대한 사전지식과 관여도가 소비자의 신념 불일치와 그 감소효과에 미치는 영향을 살펴보고자 하였다. 특히 신념체계에 대한 주관적 확률모델을 통해 논리적인 신념 불일치의 발생과 그 감소에 미치는 영향을 살펴보고자 하였다. 이에 따라 첫째, 신념 불일치의 크기는 사전지식 수준이 높은 경우에 사전지식 수준이 낮은 경우보다 더 크게 나타날 것인가? 둘째, 제품 카테고리의 관여도에 따라 사전지식이 반복측정에 의한 신념불일치 감소효과에 미치는 영향이 달라질 것인가? 라는 연구문제를 제시하였다. 연구문제 해결을 위해 두 번의 실험을 실시하였으며 그 결과는 다음과 같다. 첫째, 신념 불일치의 크기는 사전지식 수준이 낮은 경우보다 사전지식 수준이 높은 경우에 더 크게 나타났다. 둘째, 제품 카테고리의 관여도가 높은 경우에는 사전지식 수준이 높은 경우의 반복측정에 의한 신념 불일치 감소효과와 사전지식 수준이 낮은 경우의 반복측정에 의한 신념 불일치 감소효과가 차이가 없는 것으로 나타났다. 셋째, 제품 카테고리의 관여도가 낮은 경우에는 사전지식 수준이 높은 경우의 반복측정에 의한 신념 불일치 감소효과가 사전지식 수준이 낮은 경우의 반복측정에 의한 신념 불일치 감소효과보다 큰 것으로 나타났다. 결과의 의의와 시사점은 논의 부분에서 다루었다. We frequently see people who have a lot of knowledge on a subject make biased choices and decisions instead of conducting an accurate and objective information process. In other words, when people have a high level of prior knowledge, they tend to examine external information in a way that supports their prior beliefs and thoughts. When new evidence is analyzed in a direction that maintains one`s initial beliefs, the value, validity, reliability, and evaluating behavior of the evidence differs according to whether it matches one`s beliefs. In this study we examined the mechanism of biased information process due to prior knowledge on a subject. Specifically, we examined the effect of prior knowledge on the generation of systematic belief inconsistency and the size of that inconsistency through a subjective probability model. we also examined the effect of the level of involvement on the relationship between prior knowledge and belief inconsistence during that process. The term `belief inconsistency` used in this study refers to th logical inconsistency between beliefs that appear in the subjective probability model. The purpose of this study is to examine the effect of prior knowledge and involvement levels on logical belief inconsistency, which presents two research questions; a) will the level of belief inconsistency be higher when the level of prior knowledge is high than when it is low? and b) will the effect of prior knowledge on the decrease of belief inconsistency due to repeated measurement differ according to the involvement level on the product category? In order to solve these problems we conducted two experiments. Through understanding the effect of prior knowledge and involvement in a product category on the evaluation and belief formation of that product, this study will enable us to search for a way to organize persuasive messages on product attributes and benefits. The study will also contribute to the improvement of explanatory power of probability models in which the relationship between beliefs is represented in a form of subjective probabilities. The current study showed the following results. First, the average difference between the predicted belief levels according to the subjective probability model, and the level of beliefs acquired from respondents(i.e., the level of belief inconsistency which is the absolute value of the difference between the level of belief on conclusion B inferred from the reported beliefs on the four premises and the reported level of belief on conclusion B) was 0.1486, which shows that there is a significant difference(M=.1486; t (59)=9.892, p<.01). Second, the level of belief inconsistency was higher when the level of prior knowledge was high than when the level of prior knowledge was low(M(Low)=.1181 vs. M(High)=.1719; t (58)=-1.810, p<.05). That is, when the level of prior knowledge is high, the irrational cognitive tendency to analyze information according to existing beliefs and form beliefs becomes stronger, and thus the level of belief inconsistency in the probability model became higher. Third, the average value obtained by subtracting the level of belief inconsistency of the second measure from the first measure, in other words the level of decreased belief inconsistency due to repeated measurement was 0.0302(t (51)=1.847, p<.1) in high involvement product categories and 0.0270(t (60)=2.093, p<.05) in low involvement product categories, and both values were statistically significant. Therefore we confirmed the existence of the Socratic effect. Fourth, when the level of involvement in a product category was high, the decrease in belief inconsistency due to repeated measurement did not differ according to the level of prior knowledge(M(Low)=.0280 vs. M(High)=.0318; t (50)=-.116, p>.1). Fifth, when the level of involvement in a product category was low, the decrease in belief inconsistency due to repeated measurement was higher when the prior knowledge level was low than w

보상프로그램의 평가에 있어서 조절동기의 효과

나준희(June Hee Na) 한국경영학회 2008 經營學硏究 Vol.37 No.6

Recently many researchers have lively discussion on corporate rewards programs-popular marketing promotion. Prior research have mainly focused on the contents of reward (e.g. types or magnitude). That is, prior studies investigated which types of reward are more preferred by consumers, or what point is economically optimal in respect of consumer satisfaction. By the way, reward magnitude is closely correlated with its probability. It is general that trade-off relationship between outcome magnitude and probability-larger rewards for lower probability, and vice versa. Because of rare studies-although its academic and practical meaningfulness-this research investigates on relationship between rewards magnitude and its probability. Specifically, it is inquired that consumer information processing in context of judgment weighting to reward programs-which more weighted-its outcome magnitude vs. probability. This issue has important implications to corporate that have invested in rewards program. Which rewards programs are more preferred by consumers?, Who is more proper consumer for our program?, How effective are our programs? A point of these view, this research would propose a significant variable-Regulatory Focus. Promotion focus points to positive outcome, but prevention to avoiding negative outcome. So, promotion focus prefer risk-taking, but promotion prefer risk-aversion. Regulatory focus theory may have significant answers to research questions related to rewards program. That is, promotion focus relates to positive outcome (denoted to rewards magnitude), but prevention to preventing from negative outcome (denoted to (preventing from low) probability). Briefly speaking, promotion focus have fit with rewards magnitude, but prevention prevention prevention-probability fit. This hypothesis is examined by Ex. 1 and 2. Results of Ex. 1 examined our proposal. Consumers who have promotion focus preferred rewards magnitude option (=large-uncertain rewards), but prevention focus preferred probability option (=sure-small rewards.). Results of Ex. 2 reconfirmed results of Ex. 1. These results say regulatory fit remains in context of rewards program. Also, many research have suggested various regulatory fits-various domains with various variables. But this research would go a step farther-regulatory unfit issue investigated. That is, inappropriate experiential learning associated with regulatory fit may have negative influence consumer behavior consistent with regulatory fit. Specifically, prior negative experience in context of rewards program may predict reverse prior behavior. So, pre regulatory fit may moderate post regulatory fit. Ex. 3 suggested that positive experience motivates post regulatory fit-promotion to magnitude, prevention to probability-consistent with Ex. 1, 2. But, negative experience broke regulatory fit-specifically, consumers with prevention focus do not prefer probability to magnitude option. Maybe, prior negative experience would stimulate reverse action. These issue and answers of this research have practical implications to rewards program strategy. Promotional consumers are communicated with high returns-focusing, but prevention with high probability-focusing. Also, after program management is needed for controlling of negative experience and regulatory unfit effect.

NONINFORMATIVE PRIORS FOR LINEAR COMBINATION OF THE INDEPENDENT NORMAL MEANS

Kang, Sang-Gil,Kim, Dal-Ho,Lee, Woo-Dong The Korean Statistical Society 2004 Journal of the Korean Statistical Society Vol.33 No.2

In this paper, we develop the matching priors and the reference priors for linear combination of the means under the normal populations with equal variances. We prove that the matching priors are actually the second order matching priors and reveal that the second order matching priors match alternative coverage probabilities up to the second order (Mukerjee and Reid, 1999) and also, are HPD matching priors. It turns out that among all of the reference priors, one-at-a-time reference prior satisfies a second order matching criterion. Our simulation study indicates that one-at-a-time reference prior performs better than the other reference priors in terms of matching the target coverage probabilities in a frequentist sense. We compute Bayesian credible intervals for linear combination of the means based on the reference priors.

Noninformative priors for Pareto distribution

김달호,강상길,이우동 한국데이터정보과학회 2009 한국데이터정보과학회지 Vol.20 No.6

In this paper, we develop noninformative priors for two parameter Pareto distri- bution. Specially, we derive Jeffreys' prior, probability matching prior and reference prior for the parameter of interest. In our case, the probability matching prior is only a first order matching prior and there does not exist a second order matching prior. Some simulation reveals that the matching prior performs better to achieve the coverage probability. A real example is also considered.

Noninformative priors for linear combinations of normal means with unequal variances

김달호,이우동,강상길,김용구 한국통계학회 2018 Journal of the Korean Statistical Society Vol.47 No.4

For normal populations with unequal variances, we develop matching priors and reference priors for a linear combination of the means. Here, we find three second-order matching priors: a highest posterior density (HPD) matching prior, a cumulative distribution function (CDF) matching prior, and a likelihood ratio (LR) matching prior. Furthermore, we show that the reference priors are all first-order matching priors, but that they do not satisfy the second-order matching criterion that establishes the symmetry and the unimodality of the posterior under the developed priors. The results of a simulation indicate that the secondorder matching prior outperforms the reference priors in terms of matching the target coverage probabilities, in a frequentist sense. Finally, we compare the Bayesian credible intervals based on the developed priors with the confidence intervals derived from real data.

Noninformative priors for Pareto distribution

Dal Ho Kim,Sang Gil Kang,Woo Dong Lee 한국데이터정보과학회 2009 한국데이터정보과학회지 Vol.20 No.6

In this paper, we develop noninformative priors for two parameter Pareto distri-bution. Specially, we derive Jeffreys` prior, probability matching prior and reference prior for the parameter of interest. In our case, the probability matching prior is only a first order matching prior and there does not exist a second order matching prior. S ome simulation reveals that the matching prior performs better to achieve the coverage probability. A real example is also considered.

Noninformative priors for Pareto distribution

Kim, Dal-Ho,Kang, Sang-Gil,Lee, Woo-Dong The Korean Data and Information Science Society 2009 한국데이터정보과학회지 Vol.20 No.6

In this paper, we develop noninformative priors for two parameter Pareto distribution. Specially, we derive Jereys' prior, probability matching prior and reference prior for the parameter of interest. In our case, the probability matching prior is only a first order matching prior and there does not exist a second order matching prior. Some simulation reveals that the matching prior performs better to achieve the coverage probability. A real example is also considered.

Reference Priors in a Two-Way Mixed-Effects Analysis of Variance Model

장인홍,김병휘,Chang, In-Hong,Kim, Byung-Hwee The Korean Data and Information Science Society 2002 한국데이터정보과학회지 Vol.13 No.2

We first derive group ordering reference priors in a two-way mixed-effects analysis of variance (ANOVA) model. We show that posterior distributions are proper and provide marginal posterior distributions under reference priors. We also examine whether the reference priors satisfy the probability matching criterion. Finally, the reference prior satisfying the probability matching criterion is shown to be good in the sense of frequentist coverage probability of the posterior quantile.

Developing Noninformative Priors for Parallel-Line Bioassay

Kim, YeongHwa,Heo, JungEun 한국통계학회 2002 Communications for statistical applications and me Vol.9 No.2

This paper revisits parallel-line bioassay problem, from a Bayesian point of view using noninformative priors such as Jeffreys' prior, reference priors, and probability matching priors. After finding the orthogonal transformation, the class of first order and second order probability matching priors are derived. Jeffreys' prior and reference priors are derived also. Numerical examples are given to show the effectiveness of noninformative priors.

A Probability Matching Prior for the Powered Ratio of Trinomial Proportions

김주성,김혜중 한국자료분석학회 2012 Journal of the Korean Data Analysis Society Vol.14 No.6

This paper presents a probability matching prior for Bayesian analysis of a powered ratio of two trinomial proportions. The prior is derived by using Tibshirani's method. A simulation study demonstrates that the posterior probabilities of certain regions, based on the prior, approximately coincide with their coverage probabilities. For the posterior analysis of the powered ratio, this paper considers a Markov chain Monte Carlo method that facilitates the Metropolis-Hastings algorithm. It is shown that this Bayesian computation method is straightforward to specify distributionally and to implement computationally, with output really adapted for required inference summaries. The study also compares the probability matching prior with Jeffreys prior in terms of the coverage probability and the accuracy of the posterior estimate of the ratio.