Abstract
Financial Guilt, a complex emotion which influences consumers buying has wide applications in marketing. We have developed a scale to measure guilt through various sub-constructs. The dimensions of guilt have been found from papers that have dealt with guilt. Several questions representing the dimensions were framed and administered to respondents in several phases. Based on the various tests for validity, reliability and factor analyses we have developed a scale that can measure guilt. This scale will be a useful first step in testing various hypotheses that have guilt as a component. The scale is an original work and includes items developed by ourselves. Future work would include testing validity and reliability with respondents of wide geographic base establishing the scale’s relevance in several geographical regions.
Keywords: financial guilt, sub-constructs, validity, hypotheses
Introduction
Guilt is an emotion which a person experiences when he believes to have violated a certain code of conduct. This often leads to regret in the later stages when the person owns up or takes responsibility for the same. This feeling of guilt according to psychologists is difficult to get rid of. It might in some cases lead to extreme measures. This could arise in two ways- one when a person doesn’t act in a desired way and another where a person acts in an undesired way. It is driven by one’s inner conscience. Other feelings that are associated with guilt are remorse, curiosity, shame, fear. According to Sigmund Freud, ego and super ego parental imprints interact to give rise to guilt.
The lack of guilt leads to blame others, often used as a symptom to alienate psychopaths. Lack of it is treated as an abnormal state, which goes on to establish the relevance of guilt. When a person is unable to relate emotionally to others, he falls out of the moral framework according to psychologists.
The emotions are a vital part of cognitive processes. Tapping emotions can give an insight into how decisions are made – thus we can create patterns out of decisions and group individuals accordingly.
Consumer guilt is a variety of guilt and can be defined thus – “negative feeling resulting from a consumer buying decision that is transgressing against one’s ideals. The transgression can be purchasing a product, service, idea, or experience (i.e., a brand that does not abide by quality standards), or not purchasing a product”
Consumer guilt is a specialized field widely studied today. In recessionary times it becomes more important to understand consumer decision making process. Understanding consumer psyche would help companies design communications and target the required TG in a desired manner.
Psychologists have recorded beneficial reactions as after effects to guilt. For example, when a customer gets more attention from the salesman, he tends to create a liking and loyalty to the shop as not tending to repurchase from the shop would invoke guilt. This is the concept behind Customer Relationship Management – offering lower prices; buying back unsatisfactory products are a few examples.
Guilt in Marketing
Marketing has concentrated on decision making process by consumers governed by consumer rationale. Coca Cola’s “Open Happiness” campaign is an excellent example. Companies target audience based on intangibles – which studies have shown to have greater impact than utilities and features of the product put together. Emotions like love, fear, worry have been exploited by marketers. One of the emotions that has not been explored completely is GUILT.
This emotion is a fad today – Be it the clinic plus, shampoo campaign where the kid taunts the mother for not being a housewife and being unable to maintain her hair properly. Studies reveal that Barbie dolls are sold in huge numbers when bundled along with expensive formal shirts. Fathers overcome the guilt of spending huge sum on oneself by buying their daughters Barbie dolls.
A study by Dana-Nicoleta Lascu, University of South Carolina categorises guilt into four types
- Health guilt – a form of guilt arising from the fact that a consumer is unable to take care of his health. This is aroused when one is prone to unhealthy eating practices or is not able to devote time to prune his health. Healthy drinks and healthy sugar substitutes seem to use this in abundance in their communication campaigns
- Social Responsibility guilt – Social ad campaigns like the ones against drunken driving, anti- smoking campaign take this aspect into design. However, this type of guilt is more to evoke guilt and help you perform your duties towards the society in general and under privileged in particular.
- Moral guilt – Though this is a variant of Social responsibility guilt, it is different in a way that the focus of guilt is a smaller circle. It is usually one’s close family. The ads brought out by the insurance companies evoke consumer guilt that he is morally responsible for his kid’s education and wife’s survival.
- Financial Guilt – “Financial guilt is characterized by feelings of guilt that result from making purchases that are not easily justified. ‘Unneeded’ or extravagant expenditures are examples of purchases which could stimulate financial guilt. Impulse shopping or a lack of bargain shopping may also generate financial guilt feelings.”
The focus of this study is to quantify the guilt perception of a consumer. So, we aimed at creating a scale to measure financial guilt. If a consumer feels remorse/regret after a purchase it is essential to reinforce the value and quality of the product to ensure repurchase. Thus, measuring guilt would help us generalize what triggers guilt and when guilt will be experienced. This information can be used to design suitable marketing communication campaigns and increase customer lifetime value.
Literature Survey
1) Mukhopadhyay, A. (2004). Columbia U.
Unintended purchase opportunities: Conflict, choice, and consequence.
Dissertation Abstracts International Sec A: Humanities and Social Sciences, Vol 65(6-A).
Consumers spend substantially on products they had not planned to buy. Marketers exploit this fact trying to increase purchase incidence. This dissertation studies how decisions to either buy or not buy an unintended item affects responses to later tempting offers. An unintended purchase simultaneously depends on both the goal served by acquiring the product, and the goal of not spending money unnecessarily. These goals may be in conflict, which then leads to a choice about the purchase. This choice has consequences, cognitive and affective. Complex affective outcomes like happiness, pride, guilt, remorse, and anger result from these choices. These emotions influence responses to subsequent offers such that affect-consistent advertising appeals bring more favorable results.
2) Lin, Y.T., Jen, F., Catholic U., Xia, K.N., Yuan Ze U.(2009).
The Relationship between consumer guilt and product categories.
Working paper, Conference of Association of Consumer Research.
This study focuses on the construct of consumer guilt. It tries to understand what kind of product the consumer buys in order to satisfy their self-concept. Then the relationship between self-concept and consumer guilt are tested. The findings are that consumers are more likely to feel guilty when they spend on indulgences than when they spend on necessities. If the ideal-self is different from actual-self, consumers feel guilty when they buy something to satisfy their ideal-self.
3) Bennett, R. (1998).
Shame, Guilt and Responses to Non-Profit and Public Sector Ads.
International Journal of Advertising (UK), VOL 17 No 4
This article is about the difference between guilt and shame in psychological terms, being relevant for advertising research. Evoking guilt produces, according to this paper, a positive response, whereas shame more likely leads to an avoidance reaction. A set of hypotheses relating to shame, guilt and responses to advertisements are tested on 141 university students in the UK as a sample. It was found that high-guilt advertisements evoke more favourable responses than shame-intense adverts and that whereas there was no psychological resistance to guilt-intensive communication, the sample reacted negatively to shame-messages.
4) Burnett, M.S., Lunsford, D.A..(1994).
Conceptualising Guilt in Consumer Decision Making Process.
Journal of Consumer Marketing, Vol 11, Issue 3
Consumers purchase based on many emotions. Guilt is one such emotion and is currently a topic of research for marketers trying to influence this process. Focus groups, composed of a varied sample population demographic, were used to construct guilt. Four types of consumer guilt: financial, health, moral, responsibility, were identified. Consumer guilt is further classified as anticipatory and reactive, occurring in decisions to both purchase as well as not.
5) Bagozzi, R.P., Gopinath M., Nyer, P.U. (1999).
The Role of Emotions in Marketing
Journal of the Academy of Marketing Science (USA), VOL 27 No 2
Emotions are different from moods and attitudes. This paper approaches the measurement of emotions by examining effect of emotions upon behaviour. Emotions also affect the cognitive processes, goal-directed behavior and volition. Twelve areas are identified for research about effect of emotions in marketing.
6) Taylor, R.K.
Marketing Strategies: Gaining a Competitive Advantage Through the Use of Emotion
Competitiveness Review, Journal of Global Competitiveness
This article reviews theories of emotion and tries to aid in developing various advertising strategies. There are ways in which managers can link these strategies with the different emotional states of customers. Also, customer defense mechanisms are also considered and strategies to break them are listed.
7) Gangemi, A., Mancini, F.(2007).
Guilt and Focusing in Decision Making
Journal of Behavioural Decision Making (UK), VOL 20 No 1
Based on research that found relationships between emotional state, cognition and decision-making, this paper measures the effect of fear and anger in influencing positive and negative judgments. It proposes that feelings of guilt will lead to negative hypotheses. Three experiments are conducted wherein three groups of students are required to recall anger, fear or neutral emotional events, and then decide what to do with an unexpected monetary gift. In conclusion, decision strategies are affected by the person’s emotional state. Individuals in the guilt group were more focused on future guilt, and hence on relatively undesirable decision options.
8) Coulter, Pinto, Mary Beth. (1995).
Guilt Appeals in Advertising: What are Their Effects?
Journal of Applied Psychology, Vol 80(6). pp. 697-705.
This study measures consumer responses based on emotion, attitude, attributions of brands and purchase intention by changing the level of guilt induced. Sixty working mothers were taken as a sample set and proved that moderate guilt induced more felt guilt than extreme levels. An inverse relationship was established between level of guilt and attitudes toward attributions of brands.
9) Lascu, D.N., University of South Carolina. (1991).
Consumer Guilt: Examining the Potential of a New Marketing Construct
Advances in Consumer Research, Volume 18.
Even though consumer guilt is often exploited to achieve competitive advantage by marketers, enough research has not been done to document it’s effect in academic terms. Guilt can be used to persuade. This paper focuses on establishing guilt in consumers and its constructs. It develops an inventory for future usage by researchers trying to study guilt and its effect on advertisement response, sales and repurchase.
Scale development
The aim of the paper as discussed above was to develop a scale to measure guilt. We went ahead with the conventional steps of scale development.
The first step was to identify dimensions. After studying in depth the above literature, we concluded that consumer financial guilt includes the following dimensions:
a) Personality traits –
Personality traits include self esteem and locus of control. They are defined as follows:
Self-Esteem:
High self-esteem individuals think highly of themselves, are more likely to spend more on products and services that make them feel good. However, research indicates that people who rate high in self-esteem use avoidance defense mechanisms which lead them to reject threatening communications such as guilt appeals and to be more receptive to optimistic messages. Low self-esteem individuals, on the other hand, tend to use defenses which lead them to accept threatening appeals. So guilt appeals work better on them.
Locus of Control:
Individuals with an external locus of control, believing that external forces control one’s destiny, are more likely to adhere to the recommendations contained in the advertisement prescribing modalities of reducing guilt than individuals with an internal locus of control. A person with an internal locus of control does not rely on inputs from any external sources to form opinions about brand attributes.
b) Focus of guilt:
Focus of Guilt is described in terms of who is affected by the actions of the decision maker. An individual′s purchase decision may have adverse effects either on that individual or on others. It identifies whom the guilt is directed toward, because the focus of the guilt may affect the salience of the emotion. Some individuals may hold little concern for the consequences of their actions on others while some individuals may care deeply how their actions affect others. This focus determines their purchase decisions
c) Utility of product:
The utility of a product is a factor the consumer is always apprehensive about. The consumer should perceive that the value of the product is higher than the price he has paid for it i.e., the marginal utility is positive. The more the value of the product in the consumer’s perception, the lesser is the guilt he suffers from a better opportunity lost. The worse the product performs, the more the guilt that the consumer could have used the money resource for something more useful.
d) Guilt Interacting with Other Negative Emotional Appeals (Shame, Fear) –
Emotions interact with each other. One emotion may activate, amplify, or attenuate another. As seen from the research paper, inducing shame/fear will not tilt consumers in your favor while evoking guilt might prove beneficial.
After identifying dimensions, exploratory research was done and questions decided upon. We arrived at a set of 44 questions, which is mentioned under Appendix 1.
Face validity
The next step was testing face validity. This helps in viewing the questions from respondent’s point of view. This was done to eliminate misleading and confusing questions.
For the Guilt measurement survey, we designed 44 items under four dimensions. The survey was shown to 20 first year MBA students who found out the items that were confusing and double barreled. They removed 2 such items.
Content Validity
Next, content validity was tested by checking whether the items included in the scale measure the construct of interest. Content Validity was performed by administering the survey to experts that included 7 professors of the Industrial psychology, Organizational behavior, and Marketing and Research methods. It was also administered to 18 Research students and senior students of marketing domain. The specialization of these students in marketing and organizational behavior was crucial because our scale was designed for the marketers and these students had a good idea about which question is relevant for marketers. Moreover, guilt as an emotion is easily understood by the students of Organizational Behavior as they deal with emotions of the employees. The relevance of the dimensions to the scale was analysed by them. They were given the dimensions and related definitions in Step 1 and asked to mark questions as ‘relevant’, ‘relevant but not essential’ and ‘irrelevant’. They removed 15 items that did not tap the dimensions. The questions were eliminated if 50% of the people found the question irrelevant. The method used to decide on this was Lawshe’s method of determining Content Validity. A content validity ratio is calculated which is given as below for each question:
CVR= (n-N/2) / N/2
n- Number of experts who considered question relevant;
N- Total number of experts consulted.
A positive value for CVR implies that more than 50% of experts agreed to relevance of the question. The final number of questions that were considered for the survey was 27.
Phase 1: Data collection
The survey was administered to 140 candidates in the first phase through online survey tools. The respondents covered were from wide cross-section of age groups, locations and professions. A 5-point Likert scale was used to measure the responses more accurately. The responses of the Likert scale that were collected online were transferred to spreadsheets and the results were converted to numbers from 1 to 5 with appropriate adjustments to the reversal items (items that probed the absence of guilt).
Item-Total correlation
The item-total correlations of the individual items were found using correlate function of SPSS software. The 6 items that had item-total correlations less than 0.5 were dropped from the scale. The remaining 21 items had item-total correlations ranging from 0.5 to 0.72. Of these 21 items, 5 represented the category of locus of control, 5 represented focus of guilt, 6 represented utility of the product, and 5 represented Shame and Fear.
Item-Item correlation
The item-item correlations of the 21 items were found using bivariate correlation function of SPSS software. All the 21 items had item-item correlations of greater than 0.3 with 50% of the items. Hence all the items were taken into consideration.
Exploratory Factor Analysis (EFA)
After studying the above literature review in-depth, we conducted an analysis of what constructs are included in the construct of consumer guilt. (Ref: Burnett, M.S., Lunsford, D.A..(1994).
Conceptualising Guilt in Consumer Decision Making Process Journal of Consumer Marketing, Vol 11, Issue 3). Following which, we conducted an exploratory factor analysis to determine the validity of the listed constructs for consumer guilt.
Exploratory Factor Analysis was performed on the 21 items. The rotated component matrix of the items was obtained using SPSS which was analysed. It is as follows:
Rotated Component Matrix |
|||
Component |
|||
|
1 |
2 |
3 |
|
VAR00001 |
0.192226 |
0.195225 |
0.713085 |
VAR00004 |
0.788679 |
0.318717 |
0.137229 |
VAR00005 |
0.285346 |
0.244619 |
0.6489 |
VAR00006 |
0.795421 |
0.146635 |
0.314012 |
VAR00007 |
0.369396 |
0.741676 |
0.125354 |
VAR00009 |
0.373954 |
0.734327 |
0.084158 |
VAR00010 |
0.450758 |
0.178785 |
0.504564 |
VAR00011 |
0.108146 |
0.599961 |
0.340067 |
VAR00012 |
0.256011 |
0.455263 |
0.49496 |
VAR00013 |
0.404614 |
0.111261 |
0.602676 |
VAR00014 |
0.222377 |
0.820487 |
0.151313 |
VAR00015 |
0.078377 |
0.41711 |
0.631738 |
VAR00016 |
0.46313 |
0.389034 |
0.418036 |
VAR00017 |
0.686262 |
0.282411 |
0.271358 |
VAR00019 |
0.110651 |
0.650488 |
0.340389 |
VAR00021 |
0.573905 |
0.312914 |
0.367908 |
VAR00022 |
0.374771 |
0.271764 |
0.51113 |
VAR00023 |
0.475229 |
0.223969 |
0.561002 |
VAR00025 |
0.203401 |
0.626945 |
0.316887 |
VAR00026 |
0.468221 |
0.159089 |
0.552527 |
VAR00027 |
0.571195 |
0.253548 |
0.372853 |
Of these, 5 items that were cross-loaded on two factors were removed from consideration. The remaining 16 items were divided into three factors. The factors had 6, 6 and 4 items respectively. The items thus obtained were meaningful and useful. The correlations obtained in the test were used to refine the questionnaire and arrive at the final three constructs as given below:
Factor 1 – you are not affected by the ads& promos, friends’ buying behavior, friend’s perceptions or brand value but you buy for the utility of the product
Factor 2 – you’ll spend for others than on yourself and you’ll buy the product with best utility / customer service no matter how attractive the product is
Factor 3 – you’ll buy products on others’ recommendation and you are proud of the expensive products you own and you compensate family members with expensive products for the time not spent with them
The factors thus could be broadly categorized thus:
Factor1 – Locus of control
Factor 2- Focus of Guilt
Factor 3- Impulse behavior
Reliability
The reliability was tested using Cronbach’s alpha test. The Cronbach’s alpha component was found to be 0.93 using SPSS which indicates that the research conducted is reliable. This high alpha level confirms the consistency of the responses from the subjects.
Phase II – Confirmatory Factor Analysis (CFA)
Exploratory factor analysis (EFA) resulted in three factors and a set of 16 variables that influence the particular factor. In order to determine whether each of the factors can be fitted into the data set, we performed Confirmatory factor analysis (CFA). For this, we invited responses for this set of 16 survey questions and an additional question for nomological validity. The nomological validity question was -“You recently made an expensive purchase. After the purchase you saw an ad of the product. The ad claims the user to be special as he deserves the product. How important is this to you”. In order to ensure that we do not get biased responses, the set of respondents in this Phase was different from that of Phase 1. We received 122 responses for this Phase.
CFA was performed using AMOS. Regression weights were obtained for each of the common factors and its indicators. The AMOS text output is tabulated below:
We can see that the regression weights for all the variables are greater than 0.7. Also, the factor loading of the factors with the nomological validity variable is greater than 0.4. However there were 5 questions whose Regression estimates were greater than 1. These values represent the excessive loading, thus they are treated outliers. After eliminating those variables, we are left with 11 questions that correctly represent a correct fit with the given data set.
The final variables and the factors are tabulated below:
Factor 1 |
Variable 4 |
No amount of advertisement, promotions, peer influence can affect your buying of a particular brand |
|
Variable 16 |
You like to boast about the expensive products that you own |
|
|
Variable 21 |
Once you buy a product, you use it only till the brand is popular |
|
|
Variable 27 |
You will spend on expensive deodorant/body spray to avoid embarrassment when interacting with others |
|
Factor 2 |
Variable 7 |
You will only buy what you need no matter how attractive a new product is |
|
Variable 9 |
You go out specially to buy a gift for someone else and not for yourself |
|
|
Variable 11 |
You would gift your free gift coupon to someone close without regret |
|
|
Variable 19 |
You would not mind paying a product more money than its utility for brand value |
|
|
Variable 25 |
You would tend to buy a product though it is priced higher if customer service was good |
|
Factor 3 |
Variable 5 |
You go without the intention to buy but end up buying most of the products based on your other’s recommendation |
|
Variable 15 |
You enjoy other’s attention due to your possession of expensive products |
Convergent and Divergent Validity
Since all loading factors that relate the variables to the dimensions are greater than 0.7 in the questions retained convergent validity is automatically established. The AVE scores for each of the factors were calculated. The values are as follows:
Divergent Validity |
AVE |
Factor1 |
0.84 |
Factor2 |
0.82 |
Factor3 |
0.94 |
Thus, each factor adequately represents at least 84%, 82% and 94% of its corresponding variables. This is a test of discriminate validity and it further strengthens the reliability of our scale.
Results Summary
The various tests are summarized below:
No of items |
Type of analysis |
Result of Analysis |
|
Items generated |
44 |
Judgment and editing |
Generation of items |
Face validity |
38 |
Validity by colleagues |
Elimination of redundant items |
Content Validity |
27 |
Expert review |
Refining of items |
Validity test |
21 |
Item to Total correlation and Item to item correlation |
21 values correlated with item total correlation coefficient > 0.4 and item to item correlation coefficient greater than 0.3 |
Reliability test |
21 |
Cronbach analysis |
Cronbach alpha coefficient = 0.93 |
Exploratory factor analysis |
16 |
Factor analysis |
21 items are categorized under 3 factors |
Confirmatory factor analysis |
11 |
CFA using AMOS |
Confirms that each of the items in a factor can be fitted in the data set |
Convergent Validity |
11 |
Check if regression weights are greater than 0.7 |
Confirms the convergence of the data set |
Divergent Validity |
11 |
Check the AVE score of each factors |
Confirms the divergence of the data set |
Discussion
Although many research papers are available on the subject of guilt there has not been any attempt to quantify guilt as an emotion as it is complex and subjective. There are no empirical scales to measure guilt. Most of the researches conducted so far are theoretical and conceptual. This research was an attempt at developing a scale to measure consumer guilt. The scale can be used to test the various propositions that relate guilt to marketing. The results obtained in validity tests show that the scale is a reliable one. The first phase showed that scale is very consistent and stable. The high Cronbach’s alpha coefficient proves that this scale is highly reliable. Phase 2 showed that content validity holds good.
Further improvements
The scale can be further expanded to cover more dimensions. The four dimensions that we have identified will not be sufficient to measure guilt as it is a complex emotion which involves a mixture of emotions in each one. Hence the other underlying emotions that cause consumer guilt have to be found out and added as dimensions. The sub constructs like utility of a product could be further sub – classified and those also need to be identified and incorporated as dimensions. The items have to be administered to more people of different income and location to nullify the impact on the scale due to the behavior of people from particular region or income. Guilt, like other emotions is momentary – high at some point and low at some other point in life. The time when the survey is conducted will unduly influence the respondents and hence the tests have to be administered at different times to check the reliability of the scale. Social desirability bias can be nullified by administering the scale along with social desirability scales. Similarly acquiescence bias should also be nullified before finalizing the scale.
Implications
The absence of an empirical scale to measure guilt has been a major constraint so far in relating it with marketing. This scale will help marketers in finding the relationship between guilt and marketing. This scale will help throw light on the way emotions influence consumers buying behaviour. So, once relation has been established, marketers could look at tapping those emotions effectively to invoke purchase and repurchase. This would increase customer life time value.
...(download the rest of the essay above)