Recursive Model: Cognitive Learning Behavior in Online Consumers

Online purchasing decisions are online consumer behavior and are an interesting phenomenon in research. This study aims to prove the concept that online consumer purchasing decisions are influenced by cognitive learning behavior. The main theory underlying this research is consumer behavior and learning. Learning theory is used to analyze consumer learning behavior online with a mix of (crossing) learning theories of behavior and cognitive learning theory. Combination (crossing) between behavioral learning theory, cognitive is called cognitive learning behavior (cognitive learning behavior). This research is a survey. The data used are primary data, with the research instrument in the form of a questionnaire. The subjects of this study are individuals namely online consumers. Online consumers in this study are millennial generation who have made online purchases on one of the e-commerce sites in Indonesia (Matahari.mall, bukalapak, tokopedia, shopee, Zilingo) with this type of product is fashion. In this study 200 respondents were selected. The study consisted of organic stimulation of marketing on the website, online purchasing decisions, cognitive learning, experience preferences. Convenience sampling sampling technique is a sampling method where sampling is based on the availability of elements and the ease of obtaining them. Collecting data with online questionnaires and distributing questionnaires through whatsapp to respondents who are easily contacted by researchers. Data analysis methods with 1) test data quality instruments (validity and reliability), 2) Analysis of Descriptive Statistics and 3) Model Analysis with SEM. The results showed that 1) Organic stimulation of marketing on the website had a positive and significant effect on cognitive learning, 2)experience preference had a positive and significant effect on cognitive learning, 3)cognitive learning had a positive and not significant effect on online purchasing decisions, 4) experience preference had a positive and not significant effect on online purchasing decisions, 5) Organic stimulation of marketing on the website had a positive and significant effect on online purchasing decisions.


Introduction
Consumer behavior is an interesting empirical concern (Darley et al., 2010;Limbu et al., 2012;) and becomes the estuary of various management theories (Kotler, 2012). Purchasing decisions are consumer behavior that is influenced by various factors (Kotler and Keller, 2009;Park and Kim, 2003). Technology is one of the factors that influence consumer behavior (Akbar, 2014). Marketing   The study of cognitive is rooted in the psychological model of learning and information which still has wide and interesting space for online consumer behavior research because there are strategic and operational considerations (Cummins et.al., 2014). Research on psychological construction in the domain of e-commerce only scratches the surface (Kim and Lennon, 2013), has not touched on the substantial. Previous research has not explored in depth the role of cognitive learning behavior in online consumers (Rosenkran, 2010; Kang et al., 2006;Luna et al., 2002;Sundar and Kalynaraman, 2004). The limitations of previous research, direct this study to explain cognitive learning behavior in consumers online (novelty). Previous studies have not examined the cognitive learning behavior model (Cognitive Learning Behavior). The problems that will be examined are 1) what and how to Recursive Cognitive Learning Behavior in online consumers, 2) What is the role of cognitive learning behavior in online consumer purchasing decisions. The specific purpose of this study is to test and verify models that describe cognitive learning behavior (Cognitive Learning Behavior) in online consumers. The urgency of the research is to enrich the knowledge of marketing management especially consumer behavior by finding new concepts and models of "Cognitive Learning Behavior" that underlie online consumer behavior.

Literature review
In understanding online consumer purchasing decisions, the theoretical approach used is the theory of consumer behavior (Kotler and Keller, 2007) which states that purchasing decisions as part of consumer behavior are influenced by the presence of marketing stimuli, one of which is promotion or communicator. Setiadi (2008) states that marketing stimuli are any communication or physical stimuli designed to influence consumers (Lin et al., 2016). Assael (1992: 128) states that every form of communication is physical, visual, or verbal communication that can affect the response of individuals. The theory underlying online promotion marketing is the theory of Kotler and Armstrong (2008) which states that promotion is an activity that communicates the benefits of a product and persuades the target consumer to buy the product (Basuki et al., 2017). Online marketing is an application of the application of information technology and the internet in the field of marketing promotion, therefore it is closely related to the theory of technology adoption or TAM (Technology Acceptance Models) by Davis (1989) which states that an application of information systems, user acceptance of the information system strongly influenced by perceptions of ease of use and perceived usefulness (Ardyanto et al., 2015;Bechmann and Lomborg, 2013;Koufaris, 2002;Priambodo and Prabawani, 2016). Kotler and Keller (2007) require that consumer decision processes are influenced by consumer psychological factors, one of which is learning. The concept of consumer learning that is online and offline has fundamental differences. Bloom's taxonomy theory states that learning covers 3 domains, namely cognitive (mind), affective (attitude), psychomotor (skill), this learning theory applies to offline buying behavior, but in online purchasing Bloom's taxonomic theory does not apply because psychomotor cannot walk in the online domain because in the online realm there is no physical form. Therefore, in understanding the online purchasing theory used is the theory of cognitive learning (mind) or affective learning. The online marketing stimulus will affect consumers' affective abilities and affective learning processes occur. Affective learning is an ability that prioritizes feelings, emotions, and reactions that are different from reasoning (Gagne, 1977). In this affective domain, consumers will provide an emotional response from marketing stimulus (Ballantine and  And, this is where the cognitive learning process takes place. Cognitive learning theory states learning is a process that involves cognitive aspects of aspects of knowledge, reasoning, or thought. The mind of consumers will receive information about products, brands and so on, then it will be processed by the human brain. In this cognitive realm, information from online marketing will increase knowledge, understanding, application, analysis, synthesis and evaluation of consumers of a product. In the mind process consumers can recall consumer experiences when dealing with these products so that the theory of CEM (Customers Experience Management) can be involved as an approach to understanding consumer experience preferences (Meyer and Schwager,2007). In detail the theoretical approach in research can be presented in the scheme below: Based on the model in Figure 1 above, the hypotheses that can be tested are as follows Organic stimulation of marketing on the website has a significant effect on cognitive learning H2 Cognitive learning has a significant effect on online purchasing decisions H3 Experience preferences have a significant effect on cognitive learning H4 Experience preferences have a significant effect on online purchasing decisions H5 Organic stimulation of marketing on the website has a significant effect on online purchasing decisions

Research method
This research is a survey with a quantitative approach. The data used are primary data, with research instruments in the form of lift / questionnaire. The questionnaire was distributed online.

Descriptive statistics characteristics of respondents
In this study it was found that the total number of respondents was 200. The total female respondents were 71.1%. Respondents in this study were dominated by age 19-26 years at 88.6%. The most dominant type of product in online purchases is clothing at 37.8%. Descriptions of respondents supporting the survey of the Snapcart Research Institute (2018) which states that the majority of consumers shopping online in Indonesia are women with a total reaching 65%, millennial and most transactions for fashion products.  Table 3 shows that the majority of respondents (75,1%) were satisfied to make online purchases on e-commerce.  Table 4 shows that the respondent's response to the marketing stimulus has an average value of the overall indicator of 3.89 (Good). The indicator with the highest average value is the speed of information access being the most dominant indicator to explain the marketing stimulus on the website  Table 5 shows that respondents' responses to experience preferences have an average value category of the overall indicator of 3.79 (Good). The indicator with the highest average value is that past experience is the most dominant indicator that explains experience preference  Table 6 shows that respondents respond to cognitive learning, which has a good average value category in each indicator. The average value of the overall indicator is 3.79 (rather good). Knowledge involvement indicators are the most influential statements for respondents in cognitive learning (Y2). Based on table 7 shows that the responses of respondents to online purchasing decisions, which have an average value of the overall indicator of 3.63 (rather good). The indicator of product stability on online purchases is the most dominant statement in explaining online purchasing decisions. Overall the indicators in this study provide good responses. The average value of each variable also shows good results. Thus the question items on the questionnaire can be used to understand the respondents' perceptions of the subject matter in the study.  Pluzinski and Qualls, 1975). Information about the product / brand is accepted by cognition (mind). Information processing about a product / brand on consumer cognition includes the process of encoding information (encoding), then storing information (storage) and retrieving information that has been stored in consumer memory (retrival) (Tolman, 2005;Aljukhadar, 2009;Martin and Dahlen, 2005)). Preference of past experience related to consumer long-term memory relationship to a product / brand, will be revealed again in the memory (cognition) of consumers.

Data quality test
Data quality is tested by the instrument validity test (reliability) and reliability (reliability). Test Validity is a test that shows the extent to which the measuring device used is able to measure what you want to measure rather than measure the other. Validity tests have criteria that must be met. That is, loading factor is required significantly and standardized loading estimate is required 5 0.05. Table 7 below shows that all indicators meet valid criteria Reliability is a test that shows the extent to which the stability and consistency of the measuring device used, thus providing relatively consistent results if the measurement is repeated. Testing Reliability with the Cronbach Alpha Test has met the criteria if the Cronbach Alpha Test is> 0.7.  In the table above, the valuesof cr and kurtosis have been obtained in the range of -2.58 -2.58. And the value of cr on multivariate is -0.113 and is at -2.58 -2.58, which means that the data is normally distributed, so the data in this study can be analyzed using Structural Equation Modeling (SEM).

Structural equation modelling (SEM)
Analysis of structural equation modeling is used to find out the structural relationship between variables. Structural relations between variables are tested for their suitability with Goodness of fit. The results of the analysis of structural equation modeling in this study can be seen in the picture as follows: Using the Basic Model analysis, the Goodness of fit size is obtained as follows: The results of the model evaluations that are based on the Goodness of Fit criteria above generally show the conclusion that the model is quite good (marginal means still acceptable). The model is quite good, meaning the hypothesized model is in conformity with the data obtained in factual terms.

Hypothesis testing
This hypothesis test is done to find out whether or not the effect of exogenous variables on endogenous variables. This hypothesis can be accepted if the prob value is (P) ≤ 0.05. The results of this hypothesis test can be seen in the table as follows:

Coefficient of determination
The coefficient of determination in the hypothesized model was obtained 0.268 or 26.8%. Based on the coefficient of determination states that the contribution of independent variables to the dependent variable is 18% and the remaining dependent variables are influenced by other variables not examined.

Discussion
This study successfully proved that marketing stimulus has a significant effect on cognitive learning. The results of this study are consistent with the research conducted by Kotler and Keller (2007) which states that marketing stimulus will influence consumer behavior This study successfully proved that cognitive learning has a significant effect on purchasing decisions. The results of this study are consistent with the research conducted by Kotler and Keller (2007) which states that marketing stimulus will influence consumer behavior This study succeeded in proving that experience preferences had a significant effect on cognitive learning. The results of this study are consistent with the research conducted by Kotler and Keller (2007) which states that psychological consumers including consumer experience in marketing will influence consumer behavior. This study has not been successful in proving that experience preferences have a significant effect on online consumer purchasing decisions. The results of this study are consistent with the research conducted by Kotler and Keller (2007) which states that psychological consumers including consumer experience in marketing will influence consumer behavior. This study successfully proved that marketing stimulus has a significant effect on online purchasing decisions. The results of this study are consistent with the research conducted by Kotler and Keller (2007) which states that marketing stimulus will influence consumer behavior.

Conclusion
This study aims to prove the concept that online consumer purchasing decisions are strongly influenced by consumer learning factors. The main theories underlying this research are the theory of consumer behavior and learning theory. Learning theory is used to analyze online consumer behavior as a consumer learner with a mix of (crossing) learning theories of behavior and cognitive learning theory. Combination (crossing) between behavioral learning theory, cognitive is called cognitive learning behavior (cognitive learning behavior). This research is a survey. The data used are primary data, with the research instrument in the form of a questionnaire. The subjects of this study are individuals namely online consumers. Online consumers in this research are millennial generation who have made online purchases on one of the e-commerce sites in Indonesia (Matahari.mall, bukalapak, tokopedia, shoppee, Zilingo, Zalora) with this type of product is fashion. In this study 201 respondents were selected. The study consisted of independent variables (web marketing stimulus), dependent variables (online purchasing decisions), and mediating variables (cognitive learning). Convenience sampling sampling technique is a sampling method where sampling is based on the availability of elements and the ease of obtaining them. Samples are taken or selected because the sample is in the right place and time. Data analysis methods with 1) test data quality instruments (validity and reliability), 2) Analysis of Descriptive Statistics and 3) Model Analysis with SEM. The results showed that 1) Organic stimulation of marketing on the website had a positive and significant effect on cognitive learning, 2)experience preference had a positive and significant effect on cognitive learning, 3)cognitive learning had a positive and not significant effect on online purchasing decisions, 4) experience preference had a positive and not significant effect on online purchasing decisions, 5)Organic stimulation of marketing on the website had a positive and significant effect on online purchasing decisions.

Suggestion
First, this study provides recommendations for subsequent research to examine the effect of cognitive learning on online consumer purchasing decisions, because the results of the study show evidence that cognitive learning has no significant effect on purchasing decisions. Second, this study provides recommendations for subsequent research to examine the influence of experience preferences on online consumer purchasing decisions, because the results of the study show evidence that experience preferences have no significant effect on purchasing decisions Third, this study provides recommendations for the next study to examine the effect of affective learning on online consumer purchasing decisions, because research has not provided evidence of the relationship of affective learning to online consumers