Incomplete split-plot designs based on α-designs: a compromise between traditional split-plot designs and randomised complete block design
The paper shows how the α-design (also known as generalised lattice) may be used for constructing incomplete split-plot designs and describes four different methods (A, B, C and D) of construction. Intra-block efficiency factors and theoretical considerations are used to compare the methods. Based on those considerations method B was considered to be the most appropriate method for trials where tests for interaction between the two factors were important and thus this method was used and most of the paper deals with trials based on this construction method. The incomplete split-plots were superior to traditional split-plots in most cases—and the increase in efficiency of the designs can be quite large—especially for comparisons involving the whole-plot treatment. The efficiency for the comparison of the main effect of the whole-plot treatment was in most cases larger for randomized complete block design than for the incomplete split-plot design, but for other comparisons the proposed designs were in most cases more efficient than a randomized complete block design. The efficiency of the designs was compared to traditional split-plot designs and randomized complete block designs using three types of data. The three types were simulated data with known covariance structure, data from uniformity trials and data from actual trials using incomplete split-plot designs for comparing cereal varieties under different growing conditions. It is concluded that the incomplete split-plot designs may be a good alternative to traditional split-plots and a good compromise between split-plots and randomised complete blocks.
KeywordsIncomplete split-plot Construction Efficiency Uniformity trials Variety trials
The author would like to thank Hanne Østergaard and Frederica Bigongiali for accepting their data to be used. Hanne Østergaard is also thanked for valuable comments on a draft for this paper. The work was partly funded by DARCOF II, the project BAR-OF: Characteristics of spring barley varieties for organic farming.
- Bigongiali F (2009) Competition between wheat cultivar and weeds in organic agriculture. PhD thesis. Scuola Superiore Sant’Anna, Piza. 148 pp, ISBN 88-901624-6-5Google Scholar
- Box GEP, Hunter WG, Hunter JS (1978) Statistics for experimenters. Wiley, New York 653 ppGoogle Scholar
- Dorph-Petersen K (1949) Parcelfordeling i markforsøg (Distribution of plots in field trials). Tidsskr Planteavl 52:111–175Google Scholar
- Heidmann T (1988) Startkarakterisering af arealer til systemforskning: I Forsøgsarealer, måleprogram og metoder. Tidsskr Planteavl Special issue. S 1958, 89 ppGoogle Scholar
- Heidmann T (1989) Startkarakterisering af arealer til systemforskning: IV Resultater fra arealet ved Jyndevad. Tidsskr Planteavl Special issue. S 2021, 163 ppGoogle Scholar
- John PWM (1971) Statistical design and analysis of experiments. The Macmillan Company, New York 356 ppGoogle Scholar
- John JA, Eccleston JA (1986) Row-column α-designs. Biometrika 73:301–306Google Scholar
- John JA, Williams ER (1995) Cyclic and computer generated designs, 2nd edn. Chapman & Hall, London 255 ppGoogle Scholar
- Kristensen K (2003) Incomplete split-plots in variety trials—based on α-designs. Seventh working seminar on statistical methods in variety testing. Biuletyn Oceny Odmian (Cultivar Testing Bulletin) 31: 7–17Google Scholar
- Kristensen K, Ersbøll AK (1995) The use of geostatistical methods in variety trials, where some varieties are unreplicated. Biuletyn Oceny Odmian (Cultivar Testing Bulletin) 26–27:113–122Google Scholar
- Whitaker D, Williams ER, John JA (2006) CycDesigN: a package for computer generation of experimental designs. http://www.cycdesign.co.nz. Accessed 2 December 2009
- Williams ER, Talbot M (1994) ALPHA+. Experimental designs for variety trials. CSIRO/SASS, Canberra/EdinburghGoogle Scholar