Bibliographic record
Abstract
The creation of new firms is a tremendously important phenomenon. Each year hundreds of millions of people are engaged in business start-up efforts (Reynolds et al., 2005). A careful review of 87 analyses in 57 recent studies confirms earlier claims that new firms play very significant roles in employment creation, productivity growth and innovations (van Praag & Versloot, 2007). Yet, new venture creation is both under theorized and empirically under studied. Due to its emergent, elusive and nebulous nature and the fact that firms not yet in existence do not appear in any sampling frames, researchers have largely been confined to exploring the phenomenon of new venture creation via survivor-biased samples and retrospective case studies. These conditions are obvious impediments to credible, scholarly knowledge development. This is why the occurrence of an approach to the systematic, longitudinal study of representative samples of on-going new venture start-up processes is potentially an important breakthrough. The Panel Study of Entrepreneurial Dynamics (PSED) (Gartner, Shaver, Carter & Reynolds, 2004b; Reynolds, 2007) was the first full scale realization of such a study. It has established a new empirical approach that – with local variations – has been employed by several large scale parallel or subsequent studies in a range of countries including Australia, Canada, Latvia, the Netherlands, Norway, Sweden, and the US. The basic design of the research can be summarized as follows (Reynolds, in press): (an approximation of) a probability sample of on-going, early-stage business start-ups is obtained through screening phone interviews with a very large number (typically tens of thousands) of adult members of households, selected through random digit dialing. Answers to a screening questionnaire determine whether respondents are ‘nascent entrepreneurs’ (NE) or not; i.e., whether they are involved in on-going but not yet operational business start-up efforts in which they are going to be (part) owners. Qualified NEs – typically a single digit percentage – are directed to a comprehensive (20-60 minutes) interview about select aspects of the emerging venture, its owners, and their actions. Eligible cases are re-interviewed at regular intervals; typically every 6-12 months over 2 to 5 years in order to follow the process and assess outcomes. Due to their comprehensiveness (often comprising several hundred variables and multiple waves of data) it is not really possible within the space given here to provide an adequate description of contents and theoretical underpinnings of the studies. Although tables 2-4 (below) summarize some such information readers are referred to the PSED and PSED II handbooks and website, which provide comprehensive information of this kind for the publicly available US data sets (Gartner et al., 2004b; Reynolds & Curtin, 2009; www.psed.isr.umich.edu). The purpose of this review is twofold. First, we aim to assess what has been learnt so far on substantive matters from the published research using this type of empirical data. In this, we will focus on micro level issues where this line of research has made unique contributions and largely leave aside the macro issue of relative prevalence of NEs across time and space, which is equally or better dealt with through the Global Entrepreneurship Monitor (GEM) research (Bosma & Harding, 2007; Reynolds et al., 2005). Second, based on the strengths and weaknesses revealed by the research undertaken we aim to develop a set of suggestions regarding how the approach can be further developed and refined to make a greater contribution in the future.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.009 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.008 | 0.010 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.003 | 0.003 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.006 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".