Global Journal of Management and Business Research, E: Marketing, Volume 21 Issue 4

extracted information on commercial success, citation, patenting outside the country (Family size), which would also indicate the patent strength. The questionnaire also carried questions relating to the demography of the patent holder in terms of gender, age, experience, qualification and number of patents owned. The questionnaire was mailed or emailed to respondents, but due to poor initial response, rigorous follow-up through email correspondence, telephone and personal contact through visits were made. Data collection took five months. The questionnaire was first pre-tested among 32 respondents with a mixed representation of inventors from IHL's, GRI's Corporates, and Individual patent holders to ensure the questions' content, clarity, and validity. Based on the feedback of the pilot study, 1 item scale was dropped from the MO dimension, and two- item scales were dropped from the Diffusion dimension. The questionnaire was also refined by making minor changes to terminology to suit the scientific community of patent holders. The questionnaire was developed in both English and Sinhala. The findings of this study were tabulated using descriptive statistics on demography and patent details, frequency, reliability testing to ascertain the consistency and reliability of the variables, Pearson’s correlation analysis to measure the multi-collinearity and Multiple regression analysis to test the hypothesis and identify the most influencing variable contributing to the commercial success of patents. Multiple Regression was used contrary to the common belief that Logistic Regression is more appropriate than linear regression for analysis with a dichotomous dependent variable. Past comparison studies between the linear and logistic regression have shown near-identical outcomes (Hellevik, 2007). Furthermore, ANOVA has been proved to be robust to show relevant results if the data in the two samples cell proportions are equal or are between .25 and .75 and there are at least 40 degrees of freedom (Lunney 1970; D'Agostino, 1971). This study's sample data meets the condition specified by Lunny (1970) and D' Agostino (1971) with cell proportion .37 and .63 ANOVA was run after transforming the data values. IV. R esults The Socio –Demography profile of the respondents who participated in the research is given in Table 2. The majority of respondents were males (86%), while the majority fell into the age groups between 35 – 55+, indicating a mature profile of inventors and researchers. Furthermore, most surveyed respondents were tertiary qualified (78%) and well experienced, with most respondents (70%) having over ten years of experience as inventors. In addition, 58% of respondents held more than one patent, and amongst them, 10% owned more than five patents. These statistics indicate a well experienced, knowledgeable and active profile of Sri Lankan patent owners. Table 2: Socio- Demography Profile of Respondents n % Gender Male 189 86 Female 31 14 Age 14 - 20 3 1 21 - 34 39 18 35 - 44 55 25 45 - 54 60 27 55+ 63 29 Education O/L 8 4 A/L 39 18 Graduate 56 32 Masters 46 20 PhD 71 26 Experience less than 5 years 15 7 5 - 10 years 52 24 More than 10 years 153 69 © 2021 Global Journals Global Journal of Management and Business Research Volume XXI Issue IV Version I Year 2021 ( ) E 36 The Impact of Market Orientation and Diffusion on Commercial Success of Patented Innovation in Sri Lanka