Previous PageTable Of ContentsNext Page

Assessing the adoption of research & extension messages about high yield wheat packages by WA farmers

R. Murray-Prior1, W. M. Sirisena2, and L.D. Martin1 and M. F. Rola-Rubzen1

1 Muresk Institute, Curtin University of Technology, Northam, WA, Australia Email r.murray-prior@curtin.edu.au
2
Faculty of Arts, University of Peradeniya, Peradeniya, Sri Lanka

Abstract

Researchers from the Department of Agriculture, Western Australia (DAWA) have conducted intensive research on wheat varieties and management practices needed to achieve high wheat yields. These have been combined into recommendations known as High Yield Packages (HYP) for the regions of the state. The key objective of a study outlined here was to assess the adoption of the HYP since about 1990 and its impact on yield and water-use efficiency. The strategy chosen was to combine: case studies of 40 wheat farmers using an in-depth qualitative survey method; a random telephone survey of 100 farmers; and graphical, statistical and regression analysis of benchmark data from Planfarm Rural Consultants from 1995 to 2004.

Three key learnings: (1) Farmers have adopted most elements of the relevant HYP although most dont know it as the HYP. Many farmers are not aware of the original source of the research underlying the HYPs. (2) There is some evidence that farmers have increased their water use efficiency since 1995. (3) A combination of research methodologies is an advantage when attempting to evaluate the effect of research and extension investment.

Key Words

adoption, wheat, high yield package, water use efficiency

Introduction

Since the early 1990’s the DAWA have been promoting, through a variety of means, a package approach for producing premium quality wheat. Five key elements are the basis of the HYP (Anderson & Garlinge 2000, p. 151): 1. Select good cropping land; 2. Use a legume rotation; 3. Control grass weeds 1 or 2 years before the wheat crop; 4. Select productive cultivars and sow them at the appropriate times; and 5. Increase nitrogen and seed rates to take advantage of improved yield potential. These principles have been refined to provide specific packages for the production of high protein, noodle and soft wheats, although the five elements remain essentially the same. Ongoing research and extension programs conducted by the Department have supported and enhanced these recommendations.

The key objective of this study was to assess the adoption levels by Western Australian wheat farmers of the HYP elements recommended by the DAWA and their consequent impact on yields and water use efficiency since 1990. It also sought to describe major expectations of wheat farmers for future research and extension, although this is not the focus of this paper.

Adoption is a complex business, influenced by many personal, social, cultural and economic factors as well as the components of the change (Pannell et al. 2005). In Western Australia alone a wide range of studies have looked at the adoption of cropping practices by farmers (e.g. Noonan and Goddard 1994; Marsh et al. 2000). However, many of these studies have focussed on single-issue adoption. In this study, the issue is more complex because, while the key elements were similar, the details of the HYP varied across the high protein, noodle and soft wheat packages as well as across regions. In addition, the problem was to assess the impact of these changes on wheat yield and water-use efficiency.

Methods

Because the evaluation occurred after the fact and was focussed on the end result, it can be classified as the first form of Owen’s (1993) five forms of evaluation; evaluation for impact assessment. As well, the evidence collected was mainly at Levels 5 (KASA), 6 (behaviour change) and 7 (end results) of Bennett’s (1979) hierarchy.

Work on the project commenced in February 2004. Having reviewed the literature available on research related to the elements of the HYPs by the DAWA, the team decided to combine both qualitative and quantitative methods of evaluation, in order to capture the main trends in the impact of research on farmers. To get a comprehensive picture of the diffusion of HYP among wheat farmers, three strategies were chosen based around components using different methodologies.

1. Case Studies of a small number of wheat farmers in Western Australia using both quantitative and in-depth qualitative survey methods

2. Telephone survey based on a randomly selected sample of 92 wheat farmers from Western Australia

3. Analysis of data on wheat production in Western Australia obtained from the Farm Business Surveys conducted by the Planfarm consulting group.

Case studies wheat farmers in Western Australia

The research team developed a detailed questionnaire to be used for the case study that included qualitative and quantitative questions. It aimed at capturing agronomic practices related to wheat farming used by West Australian farmers representing four phases, 1990, 1995, 2000 and 2003. 1990 was used as the benchmark year to see how the practices changed gradually over the years during the last decade or so, as a result of adopting the HYP introduced by DAWA. It also incorporated questions to evaluate the perceptions of farmers about these new practices and also where they sourced their information. Some questions were aimed at assessing the reliability of the information they received and how they evaluated their sources of information. The team was interested to capture and describe the reality at the field level rather than building a model of farmer behaviour based on a particular theory.

The questionnaire was tested before it was finalised and necessary changes were made in the light of the feedback received from testing before it was used in farmer interviews.

Due to limitations in time and resources, the research team decided to include a limited number of wheat farmers in the sample of case studies. The sampling method used in choosing the case studies was judgement or purposive sampling (Churchill 1995) because the research was to collect time series data. Consequently, one criterion was that they were likely to have farm records going back to around 1990. They also had to be willing to spend at least an hour being interviewed. Case study farmers were chosen from the Northern, Central, Eastern, and Southern regions. A total of 48 farmers were interviewed.

Interviews were conducted by two researchers visited each farmer at a time. Professor Sirisena was present at all 48 interviews and Dr Roy Murray-Prior and A/Professor Lionel Martin took turns to join Professor Sirisena. Since the principal researchers were involved in all the farmer interviews, it was possible to maintain the accuracy and quality of the data collected. Each interview on average took about one and half hours, but some farmers had longer discussions taking more than two and a half hours. The first part of the questionnaire was faxed to the farmers in advance since they needed to retrieve detailed information from their farm records. In some cases, this part of the questionnaire was left with the farmer to be completed later and faxed back to the researchers.

Telephone survey

The team developed a five-page telephone questionnaire that was tested with five farmers before finalisation. Dr Fay Rola-Rubzen provided the necessary training for the research assistants. A sample of 100 farmers for the telephone survey was selected randomly using stratified random sampling (Churchill 1995) to represent the Northern, Central, Eastern and Southern wheatbelt using a list of farmers obtained from a database of farmers in Western Australia. From these 94 usable replies were obtained. Since the sample was randomly selected, findings of the survey have a more general application and it is expected to complement the research findings obtained from the in-depth analysis of the 48 case studies.

Analysis of Farm Business Survey Data from Planfarm consultants

The third component of the research project was to analyse a database related to wheat production in West Australia obtained from the Planfarm Pty Ltd who collect detailed economic data from their clients. The data was obtained from Brendan Tierney a consultant in Wongan Hills who replaced names with client numbers to maintain client confidentiality. The data were cleaned for obvious errors before analysis commenced. To date only simple descriptive analysis has been conducted, although it is planned to conduct regression analysis to examine productivity and technical efficiency in wheat farming in Western Australia. The database is time series covering the period from 1990 to 2004, which permits a comparative study of the changes in wheat agronomy using this set of data and the database generated from the in-depth case studies.

Results and discussion

The main issues to be addressed in the results include adoption of the elements of the HYPs, improvements in wheat productivity and performance of the DAWA.

Knowledge of HYP

The evidence here relates to Level 5 of Bennett’s Hierarchy (Bennett 1979). Farmer responses from the telephone survey about the elements of the HYPs suggest that they are familiar with the elements of the packages, although the level of knowledge varies among farmers. However, they generally do not associate the elements with the term HYP with about half saying they have heard of the HYP. Of these 20% say they are following its elements.

Adoption of elements of HYP

The evidence for the adoption of each of the 5 HYP elements is outlined below. It provides evidence at Level 6 of Bennett’s Hierarchy (Bennett 1979).

(a) Select good cropping land

It was not possible to collect hard evidence on whether good cropping land was selected because this would have required an examination of paddocks. However, evidence from the case study survey suggests many are selecting land for cropping based on its productive potential often assessed by potential yield.

(b) Use a legume rotation

Crop rotation, being one of the components of the package, has undergone substantial changes during the last fifteen years with grain legume crops added into the traditional pasture-wheat rotation. Of farmers in the telephone survey, 47% say their wheat follows a good legume or pulse crop. This seems high, possibly due to the wording of this question, which could possibly have multiple interpretations. Evidence that wheat is less like to follow a legume comes from the Planfarm data (Figure 1) which shows a decline in the proportion of crop sown as legumes (%) from 1995 (25%) to 2004 (16%). Much of this change is correlated with the decline in sheep numbers in the grain belt of WA during the mid 1990s and the removal of pasture legumes from the rotation. It also coincides with the adoption of no-till farming techniques. This decline appears to have stabilised more recently.

Figure 1. The change in the proportion of land sown to cereal versus legumes from 1995 to 2004. Data obtained from Planfarm Farm Business Surveys (Planfarm 2005, pers. com.).

Evidence from the qualitative survey suggests farmers have a good understanding of the value of cereal - legume rotations, but are aware of the economic tradeoffs involved with incorporating pasture legumes and lupins in their rotations.

(c) Control grass weeds before wheat crop

Evidence from the qualitative and telephone surveys supports the idea that the need for this is well understood and practiced. Farmers are using a range of practices (see Table 1) to help them control weeds and to overcome problems with herbicide resistance. Around half of farmers are using pasture and crop topping, with over a third using double knock sprays prior to seeding.

Table 1. Methods of weed control used by respondents to the telephone survey.

Method of weed control

Respondents using each method (%)*

Pasture topping

52

Cultivation

27

Single knock

35

Double knock

36

Pre-emergent herbicides

79

Post-emergent herbicides

69

Crop topping

46

* n = 92. More than one answer allowed.

(d) Select productive cultivars and sow at the right time

Farmers are familiar with the new wheat varieties released by the DAWA and within a short period after their release, they have used these varieties in their wheat cropping. Most are selecting productive cultivars, with 83% using at least one recommended cultivar from the Crop Variety Sowing guide recommendations (Littlewood 2003) for their area.

Nearly all (95%) are aware of the concept of a ‘flowering window’ (Anderson & Garlinge 2000) while 51% use it. They are constantly making choices of varieties in order to catch the ‘flowering window’ by combining short, medium and long-season varieties. Most of the farmers are making an effort to bring the seeding schedule to fall within the recommended time period depending on the break of the season. Flowering windows are most useful for farmers in frost prone areas, while farmers in the north of the state where this is not an issue, are more concerned about the related issue of season length of variety because of the potential for dry finishes.

Farmers are also aware of the impact of delayed seeding on yields and are keen to start as early and finish as quickly as possible. This is a greater problem for those in the north and north-eastern parts of the grain belt because of their shorter seasons. In 2003, the average start for seeding of wheat was the 11 May and the average length of seeding was 22 days.

(e) Increase nitrogen and seed rates to take advantage of yield potential

Evidence from the Planfarm data suggests nitrogen use has increased by about 30% between 1995 and 2004, from around 35 kg/ha in 1995 to 50 kg/ha in 2004. Most of this increase occurred in the first five years and appears to have levelled off during the latter five years perhaps due to the lower rainfall experienced during that period (Figure 2).

Figure 2. Increasing N rate on wheat and influence of growing season rainfall. Data obtained from Planfarm Farm Business Surveys (Planfarm 2005, pers. com.).

Seventy-six percent of farmers have increased their seeding rates since 1990. The current seeding rate averages 65 kg/ha, but this ranges depending on the region and their climatic conditions (Table 2). Some farmers are also using increased seeding rates above 100 kg/ha to control weeds.

Table 2. Average seeding rate by region from telephone survey data.

Region

Seeding rate
(kg/ha)

Eastern

59

Midlands & Central

70

Upper Great Southern

73

Central Great Southern

70

Great Northern & Mid West

71

South Eastern Lakes

55

n = 93.

Compared to what they were doing fifteen years ago farmers have become more precise in their application of fertilisers by adjusting rates depending upon expected yield, prices, soil fertility and crop requirements. Farmers were also concerned about the quality of wheat, and they knew that adequate nitrogen was needed to achieve appropriate protein levels for certain varieties of wheat, which in turn will bring them better prices for premium quality.

Improvements in wheat productivity

A key objective of the high yield package is to increase yields, which is an End Result at Level 7 of Bennett’s Hierarchy (Bennett 1979). However, the range in rainfall between regions and years makes it difficult to use yield to determine improvements in productivity (see Table 3). There is no obvious increase in yields over the period 1995 to 2004.

Another measure of wheat productivity is water use efficiency, defined as kg/mm of growing season rainfall. This measure is not as susceptible to variation in rainfall although in some areas it tends to be higher during years of lower rainfall. Frosts which are more likely to occur in those years can be a confounding factor because they can abort grain formation.

Table 3. Average wheat yields and growing season rainfall by year. Data obtained from Planfarm Farm Business Surveys (Planfarm 2005, pers. com.).

Year

Wheat yield
(t/ha)

Growing Season Rain
(mm)

1995

2.16

273

1996

1.99

292

1998

2.10

285

1999

2.26

358

2000

1.47

174

2001

2.09

197

2002

1.28

138

2003

2.60

263

2004

1.99

212

There appears to be an increase in water use efficiency of around 0.3 kg/mm GSR per year (Figure 3). Rainfall has fluctuated over that period with 2003 and 2004 being closer to the earlier years after the low rainfall periods in 2000, 2001 and 2002. While the problem of improved efficiency in higher rainfall years may be part of the explanation for the increase in water use efficiency, the figures for 2004 and 2005 do not appear to be consistent with that being the only explanation. Further analysis of the data is required to help ascertain the key drivers of water use efficiency.

Figure 3. Trend in average wheat water use efficiency from 1995 to 2004 along with average growing season rainfall. Data obtained from Planfarm Farm Business Surveys (Planfarm 2005, pers. com.).

Performance of the DAWA

Farmer responses showed that they were using a variety of sources to obtain information about wheat farming, but most of their information came from DAWA. For example, the Crop Variety Sowing Guide (Littlewood 2003) has been the ‘bible’ for most wheat growers and has played a major role since its introduction in the early 1990s. Over three quarters of farmers in the telephone survey currently use the Crop Variety Sowing Guide and 72% use The Wheat Book (Anderson & Garlinge 2000). DAWA farm notes are also popular, with 75% using them.

The DAWA people and publications still rate highly as the main sources of information on varieties and seeding rate (Table 4) although private consultants are increasing their influence in these areas. However, much of the information used by private industry is based on DAWA trials and publications so these figures probably underestimate their influence in this as in other areas.

Table 4. Main sources of information on varieties and seeding rate used by farmers in the telephone survey.

Source

Varieties
(% of respondents)

Seed rate
(% of respondents)

DAWA people

39

23

DAWA publications

40

7

Grower groups

0

3

Private consultants

34

26

Company agronomists

20

19

Own experience

40

44

Other farmers

48

23

General media

20

4

n = 89. More than one answer allowed.

Overall, in the telephone survey, DAWA received ratings of 6.1 and 5.7 out of 10 for the impact of wheat research and extension respectively since 1990. However, in both the telephone and the case study interviews farmers commented that they perceived there was a decline in performance over that time. A common reason suggested for this was the decline in the number and expertise of DAWA staff in the regions.

Conclusion

A high proportion of farmers have adopted most elements of HYP although they do not necessarily know it as the HYP. The messages derived from DAWA research are getting out to farmers although not all farmers are aware of the original source of the research. The DAWA is fairly highly rated particularly as a source of information on wheat varieties and seeding rates.

There is evidence for an improvement in efficiency of wheat production (as measured by water use efficiency or kg/mm GSR), but more work needs to be done to ascertain the key drivers of these improvements and the percentage improvement that is occurring.

References

Anderson, W.K. & Garlinge, J.R. (2000). The wheat book: principles and practices, Bulletin 4443, Department of Agriculture, Western Australia, South Perth.

Bennett, C.F. (1979). Analyzing impacts of extension programs, US Department of Agriculture, Washington, DC.

Churchill, G.A. (1995). Marketing research :methodological foundations, 6th edn, Dryden Press, Sydney.

Marsh, S.P., Pannell, D.J. & Lindner, R.K. (2000). The impact of agricultural extension on adoption and diffusion of lupins as a new crop in Western Australia. Australian Journal of Experimental Agriculture, 40(4), 571-583.

Littlewood, N. (2003). 2004 crop variety sowing guide for Western Australia, Bulletin 4592, Department of Agriculture Western Australia, South Perth Western Australia.

Noonan, J. & Gorddard, B. (1994). Time lags to adoption: acquisition of variety information by West Australian wheat growers, Agricultural Economics, School of Agriculture University of Western Australia.

Owen, J.M. (1993). Program evaluation, forms and approaches, Allen and Unwin, St Leonards, NSW.

Pannell, D.J., Marshall, G.R., Barr, N., Curtis, A., Vanclay, F. & Wilkinson, R. (2005) Understanding and promoting adoption of conversation technologies by rural landholders. Unpublished manuscript, submitted to Australian Journal of Experimental Agriculture. Accessed 16 November from http://www.general.uwa.edu.au/u/dpannell/dp0502.htm.

Previous PageTop Of PageNext Page