Data Collection and Analysis


The purpose of the previous chapters was to establish a theoretical and methodological framework for this dissertation. As a next step, this chapter will present how empirical data was collected and analyzed. The goal of this investigation is to obtain a representative picture of entrepreneurial activity in research units, i.e. research laboratories and centers, at U.S. universities, and draw respective conclusions.


Partial Little Square Research Unit Entrepreneurial Activity Path Coefficient Entrepreneurial Orientation 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 400.
    See Churchill and Iacobucci (2002), p. 56.Google Scholar
  2. 401.
    See Böhler (1995), p. 1770.Google Scholar
  3. 402.
    See AUTM (2003), pp. 30–31.Google Scholar
  4. 403.
    See Carnegie (2001), pp. 35–41.Google Scholar
  5. 404.
    See Carnegie (2001), p. 21.Google Scholar
  6. 405.
    For a selection of other research articles dealing with biotechnology and university technology transfer, see Owen-Smith, Riccaboni, Pammolli and Powell (2002), Zucker, Darby and Armstrong (2002), Oliver (2001), Zucker, Darby and Brewer (1998), and Liebeskind, Oliver, Zucker and Brewer (1996).Google Scholar
  7. 406.
    Oliver (2004), p. 583.Google Scholar
  8. 407.
    Powell, Koput and Smith-Doerr (1996), p. 116.Google Scholar
  9. 408.
    Murray (2004), p. 693.Google Scholar
  10. 409.
    See AUTM (2003), p. 4.Google Scholar
  11. 410.
    See Churchill and Iacobucci (2002), pp. 280–281.Google Scholar
  12. 411.
    Couper (2000), p. 464.Google Scholar
  13. 412.
    See Cobanoglu, Warde and Moreo (2001). The authors tested 300 professors with different methods and found that the fastest method was fax, with an average of 4 days to respond, followed by web sur-veys with 6 days; the slowest method was mail surveys, with 16 days to respond; average response rate was 26 per cent for mail, 17 per cent for fax, and 44 per cent for web surveys.Google Scholar
  14. 413.
    Kaplowitz, Hadlock and Levine (2004), p. 94, referring further to Couper, Traugott and Lamias (2001).Google Scholar
  15. 414.
    Kaplowitz, Hadlock and Levine (2004), p. 100.Google Scholar
  16. 415.
    See Kaplowitz, Hadlock and Levine (2004), p. 98. The authors examined 19,890 students at Michi-gan State University, all having an email account and free access to the internet. Response rate for a mail questionnaire was 31.5 per cent, in contrast to the email plus reminder postcard response rate which was 28.6 per cent.Google Scholar
  17. 416.
    Kenney and Goe’;s survey consisted of a mail survey deliver directly to faculty mailboxes, followed by two follow-up letters and one e-mail query. See Kenney and Goe (2004), p. 694.Google Scholar
  18. 417.
    Kenney and Goe (2004), p. 694.Google Scholar
  19. 418.
    See Baruch (1999), pp. 430–431.Google Scholar
  20. 419.
    See Glick, Huber, Miller, Doty and Sutcliffe (1990), p. 303. See also Seidler (1974) on informant bias.Google Scholar
  21. 420.
    Lyon, Lumpkin and Dess (2000), p. 1058.Google Scholar
  22. 421.
    See Chandler and Hanks (1993), p.Google Scholar
  23. 422.
    See Churchill and Iacobucci (2002), p. 348.Google Scholar
  24. 423.
    See Osgood, Suci and Tannenbaum (1957)Google Scholar
  25. 424.
    See Mindak (1961), pp. 28–29.Google Scholar
  26. 425.
    See Garland (1990) for a more detailed analysis of semantic differentials.Google Scholar
  27. 426.
    Miller and Friesen (1982), e.g., in their question 73 about product innovation, p. 24, use semantic differentials and differentiate between strong emphasis on true and tried products, and strong emphasis on technological leadership and innovation. However, there is an argument that a company can be both focusing on tried products, and being a technological leader.Google Scholar
  28. 427.
    For example, the first two questions of the “risk taking” construct in this study investigate the com-mitment to risky projects, and the carefulness of allocation resource into uncertain projects. These questions are not fully opposed to each another, given that one can be careful with resources, but never-theless commit to risky projects, well knowing about the risk and adjusting for it. On the other hand, these constructs control for each other in a way that one would expect opposing directions.Google Scholar
  29. 428.
    This scale was first proposed and named after Likert (1932).Google Scholar
  30. 429.
    The author would like to thank Leandro Saita of Stanford Graduate School of Business, for his comments during the Stanford University Social Science Statistical Support Group sessions, and An-drew Nelson of the Stanford University Management Science and Engineering Department for his helpful comments.Google Scholar
  31. 430.
    The author is grateful for the participation of Denise Chan, Nadja Dornhöfer, Rachel Freiberg, Ester Hammond, and Fiona Kaper, all researchers at Amato Giaccia’;s Laboratory for Radiation Biology at Stanford University, in testing and commenting on the survey.Google Scholar
  32. 431.
    The author would like to thank Andreas Kessell and Stephan Hungeling of RWTH Aachen Univer-sity for their useful comments.Google Scholar
  33. 432.
    See Kenney and Goe (2004), who report a response rate of 13.4 per cent in their study.Google Scholar
  34. 433.
    Bortz and Döring (2002), p. 400.Google Scholar
  35. 434.
    Armstrong and Overton (1977), p. 396.Google Scholar
  36. 435.
    See Kumar, Stern and Anderson (1993), p. 1634.Google Scholar
  37. 436.
    See Chandler and Hanks (1993), p.Google Scholar
  38. 437.
    See Campbell (1960), p. 546.Google Scholar
  39. 438.
    Peter (1981), p. 135.Google Scholar
  40. 439.
    Peter (1981), p. 135.Google Scholar
  41. 440.
    Bagozzi and Phillips (1982), p. 468.Google Scholar
  42. 441.
    See Peter (1981), p. 134.Google Scholar
  43. 442.
    See Peter (1979), p. 6.Google Scholar
  44. 443.
    Kerlinger (1973), p. 457.Google Scholar
  45. 444.
    Carmines and Zeller (1979), p. 12.Google Scholar
  46. 445.
    See Bollen and Ting (2000).Google Scholar
  47. 446.
    See Peter (1979), p. 6.Google Scholar
  48. 447.
    See Carmines and Zeller (1979), p. 27.Google Scholar
  49. 448.
    Hulland (1999), p. 198.Google Scholar
  50. 449.
    Cronbach (1951).Google Scholar
  51. 450.
    See Nunnally (1978).Google Scholar
  52. 451.
    See Chin (1998), p. 320.Google Scholar
  53. 452.
    See Chin (1998), p. 321.Google Scholar
  54. 453.
    See Carmines and Zeller (1979). p. 20.Google Scholar
  55. 454.
    Rungtusanatham (1998), p. 11.Google Scholar
  56. 455.
    Kerlinger (1973), p. 458.Google Scholar
  57. 456.
    See Churchill (1979), p. 67.Google Scholar
  58. 457.
    See Cohen (1960), p. 37.Google Scholar
  59. 458.
    Bagozzi and Phillips (1982), p. 469.Google Scholar
  60. 459.
    Hulland (1999), p. 199.Google Scholar
  61. 460.
    See Hulland (1999), p. 200.Google Scholar
  62. 461.
    Peter (1981), p. 135.Google Scholar
  63. 462.
    See Bagozzi (1981b), p. 327.Google Scholar
  64. 463.
    See Table 19.Google Scholar
  65. 464.
    See Table 20.Google Scholar
  66. 465.
    See Table 21.Google Scholar
  67. 466.
    See Table 22.Google Scholar
  68. 467.
    See Table 23.Google Scholar
  69. 468.
    See Table 24.Google Scholar
  70. 469.
    See Table 25.Google Scholar
  71. 470.
    See Lumpkin and Dess (1996), p. 149.Google Scholar
  72. 471.
    Geisser 1975, p. 320.Google Scholar
  73. 472.
    See Chin (1998)Google Scholar
  74. 473.
    Omission distance is 101.Google Scholar

Copyright information

© Betriebswirtschaftlicher Verlag Dr. Th. Gabler | GWV Fachverlage GmbH, Wiesbaden 2008

Personalised recommendations