Product Instructions As a Means of Fulfilling Consumers= Usage Goals

ABSTRACT - A set of three studies examined consumer’s intended and actual usage behavior when the congruency between goals and product usage instructions varied. Two goals when using products were manipulated - maximizing the outcome (benefits) from product use and minimizing the effort expended to use the product. Two straightforward forms of product instructions, commonly found on consumer products, were manipulated - one that provided a range of amounts to use (congruent with the effort minimization goal) and one that specified ideal amounts to use for particular task sizes (congruent with the outcome maximization goal). The studies examined the means of attaining one of two goals when instructions were either congruent or incongruent with the task goals. The experimental results revealed that the goal of maximizing the outcome from product usage lead to different notions regarding product preference, intentions, and the actual amount used compared to the goal of minimizing effort depending on the type of product instructions. The results of Study 2 found that consumers intended to deviate from instructions when goals and instructions were incongruent. Consumers in Study 3 deviated from instructions by using more than the instructed amount replicating the reported intentions to deviate by Study 2 consumers. Thus, this research demonstrated that actual behavior was consistent with intended behavior.

Citation:

Ingrid M. Martin and Valerie S. Folkes (2001) ,"Product Instructions As a Means of Fulfilling Consumers= Usage Goals", in E - European Advances in Consumer Research Volume 5, eds. Andrea Groeppel-Klien and Frank-Rudolf Esch, Provo, UT : Association for Consumer Research, Pages: 215-221.

European Advances in Consumer Research Volume 5, 2001      Pages 215-221

PRODUCT INSTRUCTIONS AS A MEANS OF FULFILLING CONSUMERS= USAGE GOALS

Ingrid M. Martin, California State University-Long Beach, U.S.A.

Valerie S. Folkes, University of Southern California, U.S.A.

[The authors appreciate the helpful comments of Marty Roth, C.W. Park and Niraj Dawar.]

ABSTRACT -

A set of three studies examined consumer’s intended and actual usage behavior when the congruency between goals and product usage instructions varied. Two goals when using products were manipulated - maximizing the outcome (benefits) from product use and minimizing the effort expended to use the product. Two straightforward forms of product instructions, commonly found on consumer products, were manipulated - one that provided a range of amounts to use (congruent with the effort minimization goal) and one that specified ideal amounts to use for particular task sizes (congruent with the outcome maximization goal). The studies examined the means of attaining one of two goals when instructions were either congruent or incongruent with the task goals. The experimental results revealed that the goal of maximizing the outcome from product usage lead to different notions regarding product preference, intentions, and the actual amount used compared to the goal of minimizing effort depending on the type of product instructions. The results of Study 2 found that consumers intended to deviate from instructions when goals and instructions were incongruent. Consumers in Study 3 deviated from instructions by using more than the instructed amount replicating the reported intentions to deviate by Study 2 consumers. Thus, this research demonstrated that actual behavior was consistent with intended behavior.

One important implication of this research is that consumers actively customize products in a purposeful way so that the product better fits their needs (usage goals). This demonstrates that consumers do not passively accept marketer’s mandates about how to use or not to use products. This stream of research is consistent with the warnings literature but instead of focusing on risky or toxic products it uses a common household product.

As consumers we seek product usage information that will ensure that the product will provide the benefits that we value. Firms often expend considerable resources to identify how consumers should use products and to persuade them to comply with their recommendations. For example, clothing manufacturers provide labels explaining how to clean clothing, prepared food manufacturers offer directions on packaging specifying how to combine ingredients, and pharmaceutical manufacturers identify steps for using drugs in disclosures enclosed with the medicine. As consumers we constantly face decisions on whether to follow the mandated instructions or to customize those instructions to attain the benefits that we want from product use. Firms are faced with the question of whether they should provide product usage information that will ensure that consumers attain these salient benefits and how that information should be communicated.

Instructions tend to be product specific information that vary across brands and even models within a product class. Because they convey information about procedures for using the specific brand, one would expect product users to value and adhere to instructions. However, it is well documented that people often do not comply. For example, large-scale studies of patients who are prescribed medical regimens show compliance rates range from as low as 3% to as high as 95% (Maatman, et al., 1997; Pepin, et al., 1996). Compliance is important to marketers not just when health and safety are involved but more generally because products used as intended by the manufacturer are more likely to meet the expectations created by marketers.

Research shows that people frequently deviate because they cannot comprehend instructions, find it difficult to recall a complex series of required steps or lack the time and materials for proper execution (Celuch, Lust, & Showers, 1992; Mustard & Harris, 1989; Wright, 1981). The implication is that to attain compliance one must have simple instructions, be given sufficient time to complete the task, and have the necessary implements (Wright, Creighton, & Threlsall, 1982).

Our approach to compliance with instructions emphasizes motivation when using the product. Thus, we examine peoples’ deviations from the kinds of simple, straightforward instructions commonly found on consumer products when they have read the instructions and have the ability to comply with them. We propose that compliance will vary depending on an individual’s goals, the form of product instructions, and the relationship between these two factors. A goal approach provides the theoretical grounding for understanding consumers’ motivations to comply with marketer communications concerning product usage and the resulting product benefits. Hence, this research provides a broadened understanding of a common and fundamental but somewhat atheoretical notion - that of consumer benefits.

Our research focuses on two important usage goals B maximizing outcomes from using the product and minimizing one’s effort while using the product. The importance of these goals is indicated in the pervasiveness of appeals in the marketplace that emphasize maximizing outcomes from product use on the one hand and minimizing effort on the other hand (Beach & Mitchell 1978). These same goals have been examined in the consumer decision making literature (effort minimization in brand selection and making the best choice among brands, Bettman, et al. 1998). They have been shown to be negatively correlated such that when the importance of one increases, the other decreases (Bettman, Luce, & Payne 1998). For example, when a consumer’s goal is to prepare a gourmet meal, the person is unlikely to concurrently hold as a goal having a very easy and quick meal to prepare (Martin & Folkes, 2001).

We examine people’s behavior when the means to attain their goals are readily accessible and contrast it to when the means are incongruent with the goal that they wish to attain. We compare behaviors of individuals with an outcome goal who use a product with instructions that fit with the goal to those with a mastery goal that use a product with instructions that fit with an effort minimization goal, and vice versa. Three studies focus on the effects of the fit between the usage goal and instructional format. When instructions do not fit with a consumer’s goal, the individual is more likely to deviate from even simple, straightforward instructions. Study 1 looks at whether products with instructions designed to achieve consumers’ usage goals will affect their preferences.

STUDY ONE

A consideration of the means to achieve a goal may provide some insight into why goals are incompatible. The hierarchical nature of goals has been represented in various ways in the goals literature but probably the most widely accepted structure is based on the concept of subgoals as the means to achieve higher order goals (e.g., Brewer & Dupree, 1983). For example, the way a set of product instructions are written can provide the "in-order-to" relationships that guides the achievement of a higher level goal (e.g., minimize effort required to use the product). In essence, this "in-order-to" relationship guides goal accomplishment (Lichtenstein & Brewer 1980).

Consumers are likely to have developed some generalized expectations about the means of achieving outcome maximization and effort minimization (Bettman, et al. 1998). These expectations are likely to include aspects of the product instructions. The type of instructions that consumers perceive as the means to achieve the goal of maximizing product performance may be different from those to achieve the goal of minimizing effort. They should differ in the extent to which they constrain usage. When instructions provide a means to achieve a goal, an individual should prefer that product to one that has instructions which do not provide the optimal means to achieve the goal.

Precise Instructions. People perceive that optimizing performance entails more constraints and restrictions as a means to achieve one’s goal as compared to when performance standards are more lax (Heider, 1958). Some instructions are quite constrained in their directions for product usage. An example of this type of instructions is one that guides the use of houseplant fertilizer by stating: "use 2 cup of fertilizer for a plant less than one foot tall in height and use 1 cup of fertilizer for a plant that is one foot tall or more in height". Precise instruction clearly specify the ideal points for the amount of a product to be used and so fits with the goal of outcome maximization. Hence, instructions that narrowly specify the way to use a product and allow less flexibility in use are likely to be adhered to by an individual with an outcome maximization goal.

Range Instructions. An effort minimization goal suggests a means that is less constrained than an outcome maximization goal. Instructions should require less precision on the part of the consumer. Range instructions fulfill this criterion by providing a more open-ended process. For example, instructions that provide a means of minimizing the effort to use houseplant fertilizer might state: "use 1/2 to 1 cup of fertilizer per plant". This type of instructions provides some flexibility in selecting how much of the product to use and fits better with an effort minimization goal. The instructions allow the individual with the effort minimization goal to be less precise in measuring out an amount of fertilizer. The individual could select 2 cup, 2/3 cup, : cup, or some other amount without a focus on any exact amount unlike the precise instructions.

In addition, the lack of task size information in the range instructions can be interpreted by the individual as not being important. In other words, the individual does not have to extend much effort to measure and use the product since that information is not provided in the instructions. By providing the task size information, the precise instructions draw attention to the importance of expending effort in this process, thus, ensuring "healthy, lush plants". Precise instructions are hypothesized to be the means to achieve the outcome maximization goal and the range instructions are hypothesized to be the means to achieve the effort minimization goal. Preferences for a fit between goals and sub goals should be reflected in product preferences (Guttman & Reynolds 1988).

Method

A questionnaire was given to 99 undergraduate students from three universities. The questionnaire described a houseplant fertilizer and asked subjects to indicate which of two brands they would purchase given each of the two usage goals. Half (49) the subjects were randomly assigned to the effort minimization condition. They were told that their usage goal or benefit sought was to have "a fertilizer that is as easy as possible to use" (effort minimization). The remainder (50) was assigned to the outcome maximization condition that stated "to have lush, green and healthy plants". The two brands differed only in the type of instructions (precise vs. range). The precise instructions stated to"use 1\2 cup of fertilizer for a plant less than 1 foot tall and 1 cup of fertilizer for a plant one foot tall and greater" and range instructions stated to "use 1\2 cup to 1 cup of fertilizer per plant".

Product preference was obtained followed by perceptions of goal congruency. Subjects were asked to rate both the range and the precise instructions on three 7-point scales for a total of six items. They indicated how well each type of instructions provided a means for the individual to achieve the goal, how well each ensured that the individual achieves the goals, and how well the instructions fit with the goal (1=not at all and 7=very much). The scales were collapsed into a composite measure for the goal congruent and incongruent brands (a=.86 & .87, respectively).

Results and Discussion

Given the goal of effort minimization, 82% (40 out of 49) selected the brand with the range instructions. Given the goal of outcome maximization, 90% (45 out of 50) selected the brand with the precise instructions. A difference of means test was conducted within each usage goal condition to test whether there was a difference in perception of fit between the congruent and the incongruent set of instructions. The results of the Z-test confirmed that the range instructions were perceived as the better means to attain the effort minimization goal (M=4.51 vs. 2.76, Z₄₇ =7.23, p<.0001). In the outcome maximization condition, subjects rated the precise instructions as the better means to achieve the outcome maximization goal (M=5.07 vs. 2.89, Z₄₈ = 10.39, p<.0001). In sum, the set of instructions that were the perceived means of achieving an outcome maximization goal and an effort minimization goal are different, with each set of instructions being incompatible with achieving the other goal. In addition, the product with the congruent set of instructions and usage goal was preferred to the other product with the incongruent instructions and usage goal. The next step was to determine whether this preference for congruent instructions and goals would impact consumers intended compliance with product instructions.

STUDY TWO

Although people may prefer brands with instructions that are congruent with their goals, they may be faced with incongruent instructions for a variety of reasons (e.g., the existence of uniform instruction types in a product category, legally mandated instructional formats, the individual’s goal changes over time, the instructions are not visible prior to purchase). Studies Two and Three contrast consumers’ responses when instructions are congruent with the individual’s usage goal to the situation when the instructions are incongruent with one’s usage goal. Study 2 focused on consumers’ intentions to comply with instructions and Study 3 went the next step and focused on their actual usage behaviors. The hierarchical structure of goals suggests that goal should guide behavior so that the consumer pursues the accessible course of action more closely when it is congruent with one’s goal than when it is perceived as incongruent (Lichtenstein & Brewer 1980).

In addition, we examined other ways in which the product usage instructions are perceived to differ for the two goals. The goal of outcome maximization entails both more physical and cognitive effort than does the goal of effort minimization (Chaiken, Giner-Sorolla, & Chen, 1996). Ideal outcomes are perceived to require the identification of ideal points of individual causes of outcomes and maintenance of these ideal points in the ensuing application. A focus on the physical effort involved in product usage (e.g., using measuring implements, precision in measuring amounts) as well as cognitive effort (e.g., assessments of plant health and product effectiveness) should be more apparent with the outcome maximization goal as compared to the effort minimization goal.

TABLE 1

MEAN RATINGS FOR GOAL AND INSTRUCTION CONDITIONS IN STUDY TWO

In contrast, the means to achieve effort minimization involves reducing the complexity of the decision making process (e.g., reducing the amount of information sought and used when deciding how much of the product to apply) (Chaiken, et al., 1996). It also involves reducing the number and intensity of actions needed to execute the decision (e.g., skipping prescribed steps, decreased attentiveness to the task). For example, a homeowner applying fertilizer may have as a usage goal to obtain the greenest and healthiest lawn possible (outcome maximization goal). The homeowner who assigns the same task to a teenager in the family might find the teen’s goal in applying the fertilizer is to expend as little effort as possible (effort minimization goal). Thus, intentions regarding the means of using products should differ depending on one’s goal. In sum, fit between usage goals and product instructions was tested to determine its influence on the intended compliance with instructions. Goals (but not the fit with instructions) were predicted to influence exertion of cognitive and physical effort.

Method

Subjects were 143 undergraduate students from two universities. Students came in groups of six to eight persons to a small room where they were seated around a conference table. In the middle of the table were two large plants and two small plants, two one-gallon bottles filled with a green liquid fertilizer and two measuring cups. Subjects were told they were participating in a marketing research study to test a new product concept - premixed houseplant fertilizer. The experimenter, who was blind to the goal and instruction conditions, explained that the fertilizer was premixed and could be applied directly to the plants. The small plant was less than one foot tall growing in a small pot and the large plant was more than one foot tall growing in a large pot.

The two manipulated goals were consistent with the usage situation. The effort minimization goal fit with the description of the product given to subjects - the fertilizer was described as "premixed", so it should be assumed to be easy to use. The outcome maximization goal C use of the product would ensure lush, healthy plants C also fit with the use of houseplant fertilizer in general. Thus, the situation presented to subjects was not in itself incongruent with either goal. Subjects read the product instructions and completed a survey asking them to respond to a series of 7-point rating scales. The two scales of degree of intended compliance to instructions were "how likely would you be to use more/less than the instructions state", anchored by 1=not very likely and 7=very likely.

Additional measures were collected to confirm subjects’ beliefs about the physical and cognitive effort needed to accomplish their goal. Three measures of physical effort included "how important is it to be precise about the amount of fertilizer to use", "how important is it to use a measuring cup", and "how careful do you need to be when pouring the fertilizer". The first two scales were anchored by 1=not at all important and 7=very important, and the third scale was anchored by 1=not at all careful and 7=very careful. Since these three measures were highly correlated, they were combined into a composite measure of physical effort (a=.73). Three measures of cognitive effort included "to what extent would you take into account how healthy the plants are before applying the fertilizer" (anchored by 1=not much and 7=very much), "would you think a lot about the optimal amount of fertilizer to use before putting it on the plants" (anchored by 1=not at all likely and 7=very likely), and "how interested would you be in the effect of the fertilizer on the plants" (anchored by 1=not at all interested and 7=very interested). The measures were highly correlated and combined into a composite measure of cognitive effort (a=.81).

Results

The measures of intended compliance, physical effort, and cognitive effort were analyzed using a 2 by 2 ANOVA with two usage goals (outcome maximization vs. effort minimization) and two sets of product instructions (precise vs. range instructions). The analysis reveals an interaction between goals and instruction type for one of the compliance measures (F_1,139=15.56, p<.001). When subjects were asked whether they would use more than the instructions specify, those in the incongruent conditions (outcome maximization-range and effort minimization-precise) were more likely to use more as compared to those in the congruent conditions (outcome maximization-precise and effort minimization-range) (F_1,139=13.99, p<.001 and F_1,139=21.03, p<001, respectively). Subjects do not intend to use less than the instructions specify, as indicated by the low cell means for the "underdose" measure in Table One.

Analysis of variance revealed a significant main effect for the usage goals for the composite measures of physical effort and cognitive effort (see Table One). Subjects in the outcome maximization condition indicate they would follow a different procedure than subjects in the effort minimization condition. As expected, the cell means show that physical and cognitive effort exerted was greater for outcome maximization subjects compared to effort minimization subjects (M=5.21 v 3.75, F_1,139=16.81 and M=5.59 v 2.57, F_1,139=14.63, respectively).

The goal main effect is qualified by an interaction effect for cognitive effort (F_1,139=8.46, p<.01). Contrasts reveal that more cognitive effort would be expended when outcome maximization goal subjects were given precise instructions than when they were given range instructions, (M=5.62 v. 4.80, F_1,139=11.54, p<.01). Nevertheless, differences across usage goals remain. Those individuals with the outcome maximization goal were more likely to put more cognitive effort into the usage process. This is seen in the comparison between the outcome maximization-precise instruction condition with the effort minimization-precise instruction condition and the outcome maximization-range instruction condition with effort minimization-range instruction condition (F_1,139=15.09 & F_1,139=8.10, respectively).

Discussion

The means to achieve an outcome maximization goal is perceived as requiring greater cognitive and physical effort than does an effort minimization goal. One might expect that this difference would influence compliance with instructions. Compliance typically requires physical effort (measuring amounts, careful pouring), which suggests that outcome maximization subjects would be more likely to comply with instructions compared to effort minimization goal subjects. On the other hand, deviation from instructions can require cognitive effort (e.g., generating hypotheses about the effect of overuse or underuse might be), which suggests that outcome maximization subjects are more likely to deviate from instructions compared to effort minimization subjects.

Had we not provided subjects with different types of instructions, we might have concluded simply that goals influence compliance for different reasons. However, the effect of goals on compliance is more complex. Instructions provide information in a format that is either congruent or incongruent with individuals’ usage goals. Compliance with instructions is greater when the instructions are perceived as a means to achieve the usage goals. Using more than the instructed amount appears to serve the goal of outcome maximization when products provide range instructions and to serve the goal of effort minimization when products provide precise instructions - goal incongruent instructions. Although the results are largely consistent with the predictions, they describe the manner in which actions are executed and not the outcome of usage (the amount used). It is unclear whether subjects would actually use a different amount than that prescribed by the usage instructions and this is addressed in Study 3.

STUDY THREE

Study 3 replicated Study 2 subjects’ intentions to comply with instructions by focusing on actual usage behavior rather than intended behavior. People should adhere to instructions that are congruent with their goals. When instructions are incongruent, people are likely to seek different means of pursuing their goals. A secondary purpose was to examine effects of goal congruency on confidence. Instructions that provide a means to achieve one’s goals should increase the person’s confidence in using the product. Lack of congruent information on how to achieve one’s salient usage goal should decrease one’s confidence in using the product. Certainty that one will achieve one’s usage goal increases when one has identified the optimal means to achieve the goal (Gollwitzer, 1996; Petersen & Pitz, 1988). For example, an individual who wants healthy, lush houseplants will have a higher confidence in achieving this goal with precise step-by-step instructions that provide the guidance needed than instructions that are more open-ended (range instructions).

Method

As in Study 2, the same two usage goals and the same two sets of product instructions were manipulated via experimental instructions making one of the four goal congruency conditions more salient. The task size variable was also used as a way to confirm the generality of the findings (e.g., that deviation is not limited to particular plant sizes or types). Subjects were 92 students taking marketing courses at a major university in the western U.S. The stimuli used in Study 3 were the same as those used in Study 2 in order to replicate the situation as closely as possible. There were plants of two sizes in the office but only one small and one large plant was on the table facing the subject. The plants were the same size and varieties as those used in Study 2.

The major difference between Studies 2 and 3 was that actual usage behavior was the focus in Study 3. After reading the instructions, students were told to fertilize one small and one large plant on the table with the houseplant fertilizer. After fertilizing both plants, subjects were given a questionnaire. Using live plants in this experiment provided the advantage of presenting subjects with a task in which product usage had actual consequences. All plants maintained a healthy appearance for the length of the study. No systematic differences across conditions were expected in the perceived health of the plants since plants size was randomly assigned to subjects.

Dependent measures

Two types of dependent variables were collected: behavioral measures of the amount of fertilizer poured on each plant and rating scale measures. The amount of fertilizer applied to each plant was measured by weighing the bottle of fertilizer before and after the individual poured an amount on the plant. The scale used to weigh the bottles was hidden from subjects’ view and the quantities were ascertained when subjects were not in the experimental room. Amounts were measured in pounds, with the amount indicated on the measuring cup as one half-cup weighing .23 pounds and the amount indicated on the measuring cup as one cup weighing .51 pounds The amounts marked on this standard measuring cup were slightly in error so that the quantity marked as one half cup did not in fact weigh half the quantity marked as that of a whole cup. An individual viewing the measuring cup would not detect this unintentional discrepancy.

We did not simply use the amounts stated in the instructions when developing criteria for behavior consistent with the instructions because this would hold the subjects to an unrealistically stringent standard. It is important to realize that subjects relied on a visual assessment of the quantity in the measuring cup when they used the product. The usage dependent variable was based on a more accurate measure, the weight on a scale of the quantity of fertilizer poured into a measuring cup. In the precise instruction condition in particular, measurement error could plausibly account for a degree of deviation from the exact amounts stated in the instructions. Thus, to arrive at a realistic criteria of conforming to the range instructions, we examined this interval extended from .21 to .53 lb. (a slightly broader range than the .23 to .51 lb. amount that corresponds to one half to one cup). For the precise instructions this interval extended from .21 to .25 lb. for the small plant (.23 lb. or one half cup in the instructions) and from .49 to .53 lb. for the large plant (.51 lb. or one cup in the instructions). Both instructions still maintain the same boundaries of .21 to .53 lb.

Survey measures consisted of 7-point rating scales. Subjects were asked "how confident/ certain/sure can you be about selecting the right amount of fertilizer so as to achieve the goal of outcome maximization/effort minimization" for the both plant sizes, anchored by 1=not at all and 7=very confident/certain/sure. The six measures formed two composite measures of usage confidence for the large and the small plant (a=.89 and .90, respectively).

TABLE 2

MEAN AMOUNT DEVIATED, CONFIDENCE RATINGS, AND CONTRASTS FOR STUDY THREE

Students’ perceptions of the product and the plants were also examined. They were asked "how strong do you think this fertilizer is?", anchored by 1=very weak and 7=very strong. The product was perceived as moderately strong, M=4.73, sd=.91, with no differences across conditions. The plants were also perceived similarly across conditions. The plant perception scales asked about each plant’s hardiness and fragility, anchored by 1=not at all hardy to 7=very hardy and 1=not at all fragile and 7= very fragile. The combined hardy/fragility measures were also correlated (r=.61 for the small plant and r=.58 for the large plant, p<.0001) and were formed into a composite measure of plant hardiness. Both plants were perceived as moderately hardy, M=4.41 for the small and 5.09 for the large plant with 7=very hardy/not at all fragile. In addition, there were no significant differences across task size, providing support for the generality of the results.

To rule out the possibility that expertise might also influence compliance with instructions, previous domain relevant experience was examined with subjects indicating their experience with houseplants (anchored by 1=no experience and 7=much experience), their experience with plant fertilizers, and the extent to which they had used a product similar to that in the study (anchored by 1=none at all and 7=quite a lot of similar products). As anticipated, the mean ratings suggest that subjects had little relevant experience (M = 3.47, 2.71, and 2.59, respectively). These three items form a composite measure of experience and did not differ across conditions (a = .69).

Manipulation checks

Subjects were asked to rate how well the instructions provide a means for the individual to achieve the respective usage goal, how well they ensure that the individual achieves the respective usage goal, and how well the instructions fit wit the usage goal. The three 7-point scales were anchored by 1=not at all and 7=very well (a=.90). Consistent with both Studies 1 and 2 results, there was a difference in fit with the goals and instructions (F_1,95=210.08, p<.0001). The precise instructions were evaluated as a better means to achieve the outcome maximization goal than the effort minimization goal, (M=5.32 vs. 2.77, F_1,92=124.16, p<.0001). Subjects perceived the range instructions as a better means to achieve the effort minimization goal than the outcome maximization goal (M=5.17 vs.3.38, F_1,92=114.34, p<.0001), confirming the link between instructions and goals.

Results

The actual usage amount was measured and used to calculate the average amount of deviation from instructions within each group. The results indicate that consumers are more likely to comply with instructions when the form of the instructions are perceived as congruent with a usage goal than when instructions do not fit a goal. The amount of deviation was examined with a 2 by 2 between subjects ANOVA with goals and instructions as between subjects’ factors. There was a significant interaction between goals and instructions, which highlights the impact of congruency between the goals and the instructions. Subjects deviated less from instructed amounts when the instructions were perceived as a means to achieve a usage goal, (F_1,95=47.26, p<.0001) (see Table Two). Further, subjects are more likely to use more rather than to use less than the instructed amount when given instructions that are incongruent with their salient usage goal.

The same pattern of results is also found when comparing the percent of subjects who used more fertilizer in each condition. The majority of Study 3 subjects used more when given the outcome maximization goal with the range instructions (n=17 or 71%). In contrast, a minority of subjects used more when given the same outcome maximization goal but with the precise instructions (n=7 or 30%). There was a significant difference between those subjects that used more in the congruent condition and those that used more in the incongruent condition (Z₄₅=2.81, p<.05). The congruency effect emerged for the effort minimization goal with subjects using more when given the effort minimization goal with the precise instructions (n=18 or 75%). In contrast, few subjects used more when given the effort minimization goal with the range instructions (n=11 or 46%). The difference between the congruent and incongruent conditions was significant (Z₄₆=2.06, p<.05).

Goals, instructions, and the goals’ congruency with instructions influenced subjects’ confidence in their usage behavior. Ratings were analyzed using a 2 by 2 between subjects ANOVA. Subjects expressed more confidence in their choice of amounts to use when the goal was outcome maximization rather than effort minimization (M=4.4 v. 4.0, F_1,95=11.59, p<.001), when presented with precise instructions rather than range instructions (M=4.6 v. 3.8, F_1,95=13.22, p<.0001), and when instructions were congruent with usage goals (F=73.11, p<.0001). Those with an outcome maximization goal were more confident when given precise instructions than when they were given range instructions. Those with an effort minimization goal were more confident when given range instructions rather than precise instruction.

Discussion

Study 3 results are consistent with the results of Study 1 and 2 that show subjects must reconcile the incongruency between these two types of goals. When instructions were perceived to be the optimal means to achieve a particular goal, the discrepancy between the amount used and the amount prescribed in the instructions was less than when the goals and the instructions were evaluated as incongruent. The method used in the experiment suggests that noncompliance did not arise merely because subjects did not pay attention to or did not read or comprehend the instructions. Although consumers do search for goal consistent information (Gollwitzer 1996), the procedures in Study 3 ensured that all subjects were exposed to the instructions and had plenty of time to finish the task. Also, the instructions were simple and required little effort to comprehend.

The finding that subjects in the incongruent groups deviated by using more fertilizer may be due to beliefs about the particular product category or to a general belief that using more of a product is the better direction to deviate. For example, consumers might generally assume that detecting an effect from product usage is generally more likely when using more rather than using less. Alternatively, using more than the instructed amount of a product might also be interpreted as a sign of overconfidence. Deviation implies that the individual believes that his or her means of using the product is better than that provided by experts who manufacture the product. However, the mean confidence ratings rule out overconfidence as an explanation for overdosing in the incongruent instruction conditions (see Table Two). Confidence when using the product is associated with the outcome maximization goal more than the effort minimization goal, with precise rather than range instructions, and particularly with the fit of goals with instructions. The instruction main effect reflects the added information given on task size differences in the precise instructions and not in the range instructions. Research has shown that an increase in the amount of information enhances confidence (Petersen & Pitz, 1988).

More importantly, the goal main effect appears irrelevant due to the goal by congruency interaction (see reversal in the direction of the means in Table Two). The influence of goal congruency on confidence suggests that subjects’ overuse in the incongruent conditions is regarded as somewhat tentative, consistent with Gollwitzer (1996). Thus, deviation reflects subjects’ attempts to arrive at a means of using the product that is more compatible with their goals, despite the recognition that overuse may not accomplish these goals. In sum, the congruency between usage goals and product instructions (as evidenced by the goal by instruction interaction) accounts for the differential compliance to instructions and the enhanced confidence in their usage choices.

GENERAL DISCUSSION

Taken together, these three experiments advance our understanding of postpurchase behavior, a neglected area of inquiry in marketing and consumer behavior (for an exception see Folkes, Martin, & Gupta, 1993; Wansink, 1996). Most research in consumer behavior focuses on purchase whereas most consumer behavior occurs at postpurchase. The research presented here akes a novel approach by examining ways people act upon products to achieve their goals rather than just examining the way people think about brands they might purchase or evaluate products after purchase. Consumers deviate from instructions in purposeful ways so that the experience conforms to their needs. The results demonstrate that consumers can actively customize products rather than passively accepting marketers’ mandates about how to use products. Also, this research shows a consistency between intentions and actual behaviors, a neglected area of consumer behavior research

Second, the research expands our understanding of usage goals. These two usage goals are similar to those emphasized in the consumer decision making realm (e.g., Bettman, et al.’s 1998 work comparing effort minimization and outcome maximization). Further, the foundation has been laid for the development of a taxonomy of usage goals and the identification of congruent and incongruent goals. A goal approach provides a promising path for theoretical advancement for the otherwise limited notion of consumer benefits.

The findings also provide insight into instructions as a product attribute as well as a means to attain a particular product benefit. As pervasive as instructions are, it is surprising that little thought has been given to their nature and how they guide consumption (see related research on warning labels, Stewart & Martin, 1994; Stewart, Folkes, & Martin, 2000). Instructional information can provide the starting point for use, use that can be modified by more general goal-linked knowledge.

Instructional congruency with consumers’ goals was found to influence important aspects of usage. Consumers’ health and safety often rests on compliance with instructions (e.g., Mahmoudi & Iseman, 1993). To achieve compliance, experts often recommend specific instructional features (e.g., Wright, 1981). In contrast, the research presented here suggests that merely reducing the complexity of instructions may be insufficient to elicit compliance. Instead, when marketers write instructions they need to take into account what motivates consumers and how consumers’ goals may vary across individuals and even by usage situation. Such a task may not be as formidable as it seems. Perhaps the two types of instructions examined in these studies have achieved their prevalence in the marketplace partially because of the ubiquity of these two task goals.

The findings in this research demonstrate that consumers’ deviations from instructions can be systematic and intentional rather than merely a result of random error. Lack of compliance may sometimes reflect hypothesis-testing procedures in which consumers’ search for the optimal means to attain salient goals other than those facilitated by instructions. Correcting for, or at least compensating for, systematic deviations rather than for random error should be easier for firms. Thus, a marketer who segments the market based on the benefits (usage goals) sought by a target set of customers should design instructions so that they facilitate attaining that salient goal. This can be combined with other product design strategies targeted towards a particular group. For example, marketers should offer products with range instructions and premixed solutions that hinder pouring too much to customers who want a product that is "easy to use". Alternatively, marketers should offer products with precise instructions that include warnings and admonitions to avoid using too much to customers who want a product that maximizes outcomes from use.

REFERENCES

Beach, L.F. and T.R. Mitchell (1978). "A Contingency Model for the Selection of Decision Strategies", Academy of Management Review, 3(July), 439-49.

Bettman, J.R., M.F. Luce, and J.W. Payne (1998). "Constructive Consumer Choice Processes", Journal of Consumer Research, 25(3), 187-217.

Brewer, W.F. and D.A. Dupree (1983). Use of plan schemata in the recall and recognition of goal-directed actions, Journal of Experimental Psychology: Learning, Memory, & Cognition, 9, 117-29.

Celuch, K.G., J.A. Lust, and L.S. Showers (1992). Product owner manuals: an exploratory study of nonreaders versus readers, Journal of Applied Psychology, 22, 492-507.

Chaiken, S., R. Giner-Sorolla, and S. Chen (1996). Beyond accuracy: defense and impression motives and heuristic and systematic information processing. In The Psychology of Action, New York: Guilford Press.

Folkes, V.S., I.M. Martin, and K. Gupta, K. (1993). When to say when: effects of supply on usage, Journal of Consumer Research, 20, 467-77.

Gollwitzer, P.M. (1996). The volitional benefits of planning. In The Psychology of Action, P. M. Gollwitzer and J. A. Bargh (Eds.), New York: Guilford Press.

Heider, F. (1958). The Psychology of Interpersonal Relations New York: Wiley.

Lichtenstein, E. and W.F. Brewer (1980). Memory for goal-directed events, Cognitive Psychology, 12, 412-45.

----------------------------------------