Factors That Influence E-Commerce Purchase Intentions

An Abbreviated Report

INSERT YOUR NAME HERE

Submitted as partial fulfillment of the requirements for MKT 497: Marketing Research

Thursday, November 29, 2007

1

Introduction

As part of the course requirements for MKT 497: Marketing Research at Jacksonville State University, a study was conducted with JSU students and individuals who were not JSU students in the Spring 2007 semester to examine factors that influence individuals’ intentions to engage in electronic commerce (e-commerce). Students in the class conducted face-to-face intercept interviews with potential respondents; as well as analyzing and reporting the results. This document constitutes an abbreviated report of the results of the described study.

Purpose of the Study

The management question being addressed by the study is, “what factors influence consumers in their decision to purchase e-commerce products?” This question may be addressed with the following research questions:

  1. What influence do demographic factors have upon e-commerce intentions?
  2. What influence do psychographic factors have upon e-commerce intentions?

A theoretical framework is necessary in order to address these research questions. Based on a variety of previous studies that relate to the topic at-hand, the theoretical framework, or model, depicted in Figure 1 may be used as an approach to answering the research questions.

One of the constructs in the model, Technophobia, relates to an individual’s unease with or reluctance to use technology and technologically-based products. Sinkovics, Stöttinger, Schlegelmilch, and Ram (2002), in their development of a scale to measure technophobia, found that people who suffer higher levels of technophobia are less likely to use technology or to purchase technologically-intensive products.

Figure 1

A Model of E-Commerce Client Behavior

Another model construct is Mental Intangibility, which represents the lack of clarity in one’s mind that an individual has of an item or concept. Laroche, Bergeron, and Goutaland (2001), found that as the degree of mental intangibility associated with a product increases, customers are less likely to trust and ultimately purchase the product. A third model construct is Impulse Buying Tendency. IBT has been defined as “the degree to which an individual is likely to make unintended, immediate, and unreflective purchases (i.e., impulse purchases)” (Weun, Jones, and Beatty, 1997, p. 306). It is possible, given the relative lack of physical controls, such as tangible goods filling the shopping cart, that impulse buyers may be more likely to engage in e-commerce transactions than non-impulse buyers.

Pan and Zinkhan (2006) were successful in demonstrating that on-line consumers are more likely to trust e-tailers that have clearly stated privacy policies. This lends support to the idea that privacy concerns are one of the most important factors limiting the growth of e-commerce. However, no systematic data exists that examines the extent to which consumers address these concerns by taking actions to protect themselves. Thus, a new measure, E-Privacy Concerns, will be used to examine how consumers actually do protect their online privacy.

The final two constructs in the model are Trust and Intentions. Specifically, trust relates the degree to which the customer trusts e-commerce and intentions simply are whether or not the customer believes he/she will participate in an e-commerce transaction in the future. Gefen and Straub (2004) found that as customers’ trust in electronic products increases, their intentions to purchase the products in the future increase.

Given this model and the discussion of the constructs, the following hypotheses will be tested in this study:

H / 1 / Gender / does not influence / E-Privacy Concerns.
H / 2 / Gender / does not influence / Mental Intangibility.
H / 3 / Gender / does not influence / E-Commerce Trust.
H / 4 / Gender / does not influence / Technophobia.
H / 5 / Gender / does not influence / Impulse Buying Tendency.
H / 6 / E-Buying Amt / does not influence / E-Privacy Concerns.
H / 7 / E-Buying Amt / does not influence / Mental Intangibility.
H / 8 / E-Buying Amt / does not influence / E-Commerce Trust.
H / 9 / E-Buying Amt / does not influence / Technophobia.
H / 10 / E-Buying Amt / does not influence / Impulse Buying Tendency.
H / 11 / Age / does not influence / E-Privacy Concerns.
H / 12 / Age / does not influence / Mental Intangibility.
H / 13 / Age / does not influence / E-Commerce Trust.
H / 14 / Age / does not influence / Technophobia.
H / 15 / Age / does not influence / Impulse Buying Tendency.
H / 16 / Technophobia / positively influences / Mental Intangibility.
H / 17 / Technophobia / negatively influences / E-Commerce Trust
H / 18 / E-Privacy Concerns / negatively influences / E-Commerce Trust
H / 19 / Past problems / negatively influences / E-Commerce Trust
H / 20 / E-Privacy Concerns / negatively influences / Intentions
H / 21 / E-Commerce Trust / negatively influences / Intentions
H / 22 / Age / negatively influences / Intentions
H / 23 / Impulse Buying Tendency / positively influences / Intentions

Methodology

The Data Instrument

In order to collect the data necessary to test the hypotheses, a questionnaire was developed that, in addition to a cover page that explained the purpose of the study, consisted of three major sections. The first section collected basic demographic information about the respondents. The second section of the questionnaire measured E-Privacy Concerns and the outcome of a serious problem the respondent had suffered with e-commerce transactions. The final section of the questionnaire measured the psychological variablesMental Intangibility, Trusting Disposition, E-Commerce Trust, Technophobia, Intentions, and Impulse Buying Tendency. Multi-item scales were employed since each of the psychological constructs was highly abstract in nature. An annotated version of the questionnaire may be seen in the Appendix to this report.

Each of the scales used to measure the psychological constructs, or variables, had been previously developed, tested, and validated in prior studies, with the exception of E-Privacy Concerns. A list of the scales employed, and their sources, may be seen below.

Scale / Source
Technophobia / Sinkovics, Stöttinger, Schlegelmilch, and Ram (2002)
Mental Intangibility / Laroche, Bergeron, and Goutaland (2001)
Impulse Buying Tendency / Weun, Jones, and Beatty (1997)
Trust / Plank, Reid, and Pullins (1999)
Intentions / Gefen and Straub (2004)

Collecting the Data, Checking the Data for Accuracy, and Data Entry

Data was collected on the JSU campus during the Spring 2007 semester by students in Dr. Thomas’ MKT 497: Marketing Research class. Students in the class were required to recruit and interview 15 current JSU students and non-students (at least 5 non-students, but no more than 7). Therefore, the study relied upon self-reported data, though the use of face-to-face interviews did assist in reducing the item non-response problems inherent with such data. Once the data was collected, Dr. Thomas’ checked the questionnaires for accuracy and entered the data using the process delineated below.

  1. Numbered each questionnaire (total of 404 completed questionnaires)
  2. Checked each questionnaire for missing or contradictory data
  3. Coded missing data as “9999”
  4. Defined SPSS database (please refer to attached Database Key)
  5. Created a new variable (Respondent Number)
  6. Assigned a name (SPSS limits the variable name to no more than 8 characters) and label for each question (variable)
  7. Reset decimal places for all variables from 2 to 0
  8. Coded the one Nominal variable as Gender. Coded the Ordinal variables as Year of Birth, Student Status, Full-TimeWork Status, Have Ever Worked Status, Primary Work Role, Most Current Work Industry, Ethnicity, and Greatest Amount Ever Spent E-tailing.
  9. All other variables coded as Scale (the default setting)
  10. For each Nominal/Ordinal variable, defined the value labels as appropriate for the particular question
  11. Reset “Missing” column for each variable from None to 9999
  12. Switched from Variable view to Data view
  13. Entered the data
  14. Ran Frequency Distribution on all variables. Examined the distributions for any 2-digit entries. No 2-digit entries occurred.
  15. Used Microsoft Excel’s random number generator to generate a list of 41 numbers between 1 and 404. Questionnaires with these randomly generated numbers were checked on each data entry point for data entry accuracy. No data entry errors were discovered; therefore, the assumption was made that the accuracy of the data entry could be trusted.
  16. Reverse-scored questions as necessary.
  17. Using the Year questions (question # 2), created a new variable, “Age.” Then, deleted the original Year variable from the dataset.
  18. Deleted the “Respondent” variable.

Sample Size and Description

DESCRIBE THE PROCESS BY WHICH THE MINIMUM SAMPLE SIZE REQUIRED WAS DETERMINED.

IDENTIFY THE ACTUAL SAMPLE SIZE ACHIEVED. NOTE: 404 OF THE SAMPLE CAME FROM THE FALL 2006 DATA COLLECTION.

EXPLAIN WHY A DIFFERENCE EXISTS BETWEEN THE MINIMUM SAMPLE SIZE REQUIRED AND THE ACTUAL SAMPLE SIZE ACHIEVED.

MENTION 3 CHARACTERISTICS OF THE SAMPLE (SELECT WHICHEVER ONES YOU THINK ARE APPROPRIATE OR NEEDED), THEN REFER YOUR READER TO THE ANNOTATED QUESTIONNAIRE IN THE APPENDIX FOR A FULL DESCRIPTION OF THE SAMPLE.

Results

BRIEFLY DESCRIBE YOUR CONSTRUCTION OF THE ANNOTATED QUESTIONNAIRE.

DESCRIBE HOW THE SCALES WERE CHECKED (I.E., THE RELIABILITY ASSESSMENT). DON’T FORGET TO IDENTIFY THE STANDARD USED.

DESCRIBE THE DEVELOPMENT OF YOUR SUMMATED, STANDARDIZED VARIABLES.

DESCRIBE THE TESTING OF EACH OF THE HYPOTHESES. DO NOT INTERPRET THE FINDINGS AT THIS POINT. AN EXAMPLE MAY BE SEEN BELOW.

ANOVA was used to test Hypotheses 1-5, which stated that, “Gender does not influence (H1) E-Privacy Concerns, (H2) Mental Intangibility, (H3) E-Commerce Trust, (H4) Technophobia, and (H5) Impulse Buying Tendency.” Churchill’s (1979) standard of a significance level less than or equal to 0.05 was used as a measure of whether a difference exists between the groups. Using this standard, ANOVA revealed no statistically significant differences between females and males on E-Privacy Concerns, Mental Intangibility, E-Commerce Trust, and Technophobia. Hence, the decision must be made to fail to reject Hypotheses 1-4. However, ANOVA did reveal that a statistically significant difference does exist between females and males on Impulse Buying Tendency (the results of the ANOVA test may be viewed in Table 2). Therefore, Hypothesis 5 must be rejected.

Conclusions

BASED ON YOUR FINDINGS IN THE RESULT SECTION, DRAW CONCLUSIONS ABOUT WHAT THEY MEAN. AN EXAMPLE MAY BE SEEN BELOW.

In the testing of the firstfive hypotheses, no statistically significant differences were discovered between females and males on E-Privacy Concerns, Mental Intangibility, E-Commerce Trust, and Technophobia. This means that the decision to be customer of e-commerce products, at least as it relates to these variables, is not influenced by a person’s gender. However, a statistically significant difference was discovered between females and males on Impulse Buying Tendency. Females were shown to be more likely to engage in impulse buying on-line. Thus, females and males are just as likely to have privacy concerns online, suffer from similar levels of technophobia and mental intangibility, and to have the same level of trust regarding online merchants. Finally, it does appear that females are more likely to engage in unintended purchases online.

Recommendations

BASED ON YOUR CONCLUSIONS, USE YOUR KNOWLEDGE OF MARKETING TO MAKE RECOMMENDATIONS TO M-COMMERCE MARKETING MANAGERS. IN OTHER WORDS, TELL THEM WHAT THEY SHOULD DO, GIVEN YOUR CONCLUSIONS. AN EXAMPLE MAY BE SEEN BELOW.

It was found that females and males only on Impulse Buying Tendency. Thus, it would be inappropriate to develop marketing programs that are targeted specifically for females or males, at least in regard to privacy concerns, mental intangibility, technophobia and trust. However, since females are more likely to make unintended online purchases, an e-commerce firm might develop a website that prominently featuresimpulse and convenience products that target females. Yet it would be a mistake for the online merchant to add features to its website that are designed to assist males (only) in overcoming their technophobia, as both genders exhibit essentially the same level of technophobia. Thus, if the merchant is concerned that technophobia may be an issue with its target market, website design should take into account the preferences and attitudes of both genders.

Limitations

IDENTIFY AND DESCRIBE AT LEAST 3 LIMITATIONS ASSOCIATED WITH THIS STUDY. BE SURE TO EXPLAIN WHY EACH ONE IS ACTUALLY A LIMITATION. AN EXAMPLE MAY BE SEEN BELOW.

One limitation of this study is that the data was collected by college students. While the students had been trained in how to conduct the interviews, their level of experience in doing so was low. This may have unintentionally introduced interviewer error into the results.

References

Churchill, Gilbert A. (1979), "A Paradigm for Developing Better Measures of Marketing

Constructs,” Journal of Marketing Research,16(February),64-73.

Cronbach, L.J. (1951), “Coefficient Alpha and Internal Structure of Tests,” Psychometrika,

16 (September), 297-334.

Gefen, David and Detmar W. Straub (2004), “Consumer Trust in B2C e-Commerce and the

importance of social presence: experiments in e-Products and e-Services,” Omega: The

International Journal ofManagement Science, 32, 407-424.

Laroche, Michel, Jasmin Bergeron, and Christine Goutaland (2001), “A three-dimensional scale

of intangibility,” Journal of Service Research, 4(1), 26-38.

Laroche, Michel, Jasmin Bergeron, and Christine Goutaland (2003), “How intangibility affects

perceived risk: the moderating role of knowledge and involvement,” Journal of Services

Marketing, 17(2), 122-140.

Nunnally, Jum C. (1978), Psychometric Theory, New York: McGraw-Hill.

Pan, Yue and George M. Zinkhan (2006), “Exploring the impact of online privacy disclosures on

consumer trust,” Journal of Retailing, 82 (4), 331-338.

Plank, Richard E., David A. Reid, and Ellen Bolman Pullins (1999), “Perceived Trust in Business-to-Business Sales: A New Measure,” Journal of Personal Selling & Sales

Management,” XIX(3), 61-71.

Sinkovics, Rudolf R., Barbara Stöttinger, Bodo B. Schlegelmilch, and Sundaresan Ram (2002),

“Reluctance to Use Technology-Related Products: Development of a Technophobia

Scale,” Thunderbird International Business Review, 44(4), 477-494.

Stone, Robert N. and Kjell Grønhaug (1993), “Perceived Risk: Further Considerations for the

Marketing Discipline,” European Journal ofMarketing, 27(3), 39-50.

Weun, Seungood, Michael A. Jones, and Sharon E. Beatty (1997), “A Parsimonious Scale to

Measure Impulse Buying Tendency,” In W.M. Pride and G.T. Hult (Eds.), AMA

Educator’s Proceedings: Enhancing Knowledge Development in Marketing, Chicago:

American Marketing Association, 306-307.

APPENDIX

INSERT YOUR ANNOTATED QUESTIONNAIRE HERE. IF YOU DID NOT ADD YOUR DATA TABLES TO THE END OF YOUR ANNOTATED QUESTIONNAIRE, INSERT THEM INTO THIS DOCUMENT IMMEDIATELY AFTER THE ANNOTATED QUESTIONNAIRE.

1