Table 3 shows the actual and expected (in brackets) numbers of firms with each diversification strategy type about the grand median (i.e. median size in 1992) for each nation. It can be seen that in France each diversification type had a greater proportion of firms than expected above the median. This suggests visually that the young growing firms which diversified were more likely to be larger than single business firms. Because several cells had less than 5 firms, and the general trend in all diversification types appeared the same, a chi–square test of equality of proportions was attempted to compare the proportions of all diversified with single business firms about the median. The null hypothesis of equality of medians was rejected at p < 0.005 (Yates corrected ?2 = 14.67). It was also rejected on a comparison of related (including related/vertical) with single business firms at p < 0.005 (Yates corrected ?2 = 12.36), though such internal comparison tests are less reliable. In a comparison of related (including related/vertical) with vertical and unrelated strategy types (combined to obtain expected counts in cells of at least 5), expected numbers were essentially identical to actual numbers. Thus, for France, hypothesis 3 is supported but not hypotheses 1 or 2.

Visually, Ireland and Scotland appear to follow the same pattern as France, though more weakly. Numbers of diversified firms in Ireland and Scotland were too small for the median test, except for comparing the proportions of all diversified and single business firms for Ireland and for Ireland and Scotland combined. On this test, the null hypothesis of equality of medians could not be rejected for Ireland (Yates corrected ?2 = 0.57) or for Ireland and Scotland combined (Yates corrected ?2 = 1.40). The median test is a weak test. The Mann–Whitney test is more powerful, and rejected the above null hypothesis for Ireland (T1corrected for ties = 2.086; p< 0.02) and for Scotland (T uncorrected = 142, w.95 = 130; p<0.05).

TABLE 3

Actual and (expected) numbers of young growing firms about their
median size, by diversification strategy type France

related | related/vertical | vertical | unrelated | single | row total | |

>median (85) | 40 | 4 | 7 | 3 | 89 | 143 |

(29.8) | (2) | (5.5) | (2) | (103.8) | ||

<meidan (85) |
20 | 0 | 4 | 1 | 120 | 145 |

(30.2) | (2) | (5.5) | (2) | (105.2) | ||

column total | 60 | 4 | 11 | 4 | 209 | 288 |

Ireland

related | related/vertical | vertical | unrelated | single | row total | |

>median (72) | 6 | 1 | 0 | 0 | 17 | 24 |

(3.4) | (1.5) | (0.5) | (0) | (18.6) | ||

<meidan (72) |
1 | 2 | 1 | 0 | 21 | 25 |

(3.6) | (1.5) | (0.5) | (0) | (19.4) | ||

column total | 7 | 3 | 1 | 0 | 38 | 49 |

Scotland

related | related/vertical | vertical | unrelated | single | row total | |

>median (82) | 0 | 2 | 1 | 1 | 18 | 22 |

<median (82) |
0 | 0 | 0 | 0 | 22 | 22 |

column total | 0 | 2 | 1 | 1 | 40 | 44 |

Top of page | Chapter Listing | Return to 1997 Topical Index |

© 1997 Babson College All Rights Reserved

Last Updated 06/01/98