Skip to main content

Topic D

Assessing the effects of perlite and humic acid as ameliorants for green compost when growing Ocimum basilicum (sweet basil)

Contents:

1. What type of investigation will this be?

2. Background information

3. Aim and objectives

4. Population and Sampling

5. Controls

6. Variables

7. Replication

8. Statistics

9. Influencing outcomes

10. Assumptions and Bias

11. Repeatability

12. Back to the beginning

1. What type of investigation will this be?

1

Is this an observational investigation or an experiment?

[If you would like to save a record of your answer, please type it into this Word document instead of the text box below]

We are starting out with a question or hypothesis so this is an experiment.Check your answer

Back to top

2. Background information

The undergraduate reviewed the literature on this topic.

i. The justification

Targets have been set to reduce the amount of biodegradable waste entering landfill sites. These wastes can be composted. The product is referred to as green compost. The student decided to examine the potential use of green composts.

Green compost has several less favourable qualities including its density. It may be 'lightened' by adding other materials such as perlite.

Humic acid attacks and decomposes soil minerals so releasing essential cations which are taken up by plants. Small quantities of humic acid are known to improve aspects of plant growth and affect cell elongation and root initiation (e.g. Brady & Raymond, 1996).

ii. Information about practicalities

Information was obtained as to what makes a good growing media. This information was used both when selecting potential growing media and assessing their value.

Ocimum basilicum (sweet basil) was selected as the test plant because it is easily available, it germinates well from seed and tolerates a broad range of pHs. The structure of the plant allows a broad range of measurements to be taken.

There are many ways of gathering information about plant growth. The student who carried out this investigation was training to work in the horticulture industry and therefore decided to use time to emergence (when the cotyledons were distinguishable above the soil surface), percentage emergence, fresh and dry weight of above ground material (g), the number of true leaves and leaf area.

iii. The context

Reading provided information about the impact peat extraction has on habitats that are invariably of interest for conservation and the characteristics of a successful alternative growing media.

References

BRADY, N.C. & RAYMOND, R.W. (1996). The nature and properties of soils. Prentice Hall, New Jersey.

Back to top

3. Aim and objectives

2

Draft an aim and objective(s).

[If you would like to save a record of your answer, please type it into this Word document instead of the text box below]

When this study was carried out the student had two objectives. We have included only one.

Draft Aim: Assessing the effects of perlite and humic acid as ameliorants for green compost when growing Ocimum basilicum (sweet basil).

Draft Objectives:

In this study the undergraduate investigated three objectives:

  1. Compare two green composts in relation to a peat-based media for the growth of Ocimum basilicum in terms of time to emergence (when the cotyledons were distinguishable above the soil surface), percentage emergence, fresh and dry weight of above ground material (g), the number of true leaves and leaf area after 4 weeks.
  1. The effect of perlite as a bulking agent (no perlite, 25% perlite, 50% perlite) in green compost on the growth of Ocimum basilicum as indicted by time to emergence (when the cotyledons were distinguishable above the soil surface), percentage emergence, fresh and dry weight of above ground material (g), the number of true leaves and leaf area after 4 weeks.
  1. The effect of humic acid with green compost when watered at three concentrations (no humic acid, 2.5ml l-1 water and 5ml l-1 water) on the growth of Ocimum basilicum as time to emergence (when the cotyledons were distinguishable above the soil surface), percentage emergence, fresh and dry weight of above ground material (g), the number of true leaves and leaf area after 4 weeks.

We will continue to develop the experiment relating to objective two as most elements of this design will also be found in the experiments designed to test objectives one and three.

Check your answer

Back to top

4. Population and Sampling

3

What is the statistical population? Will you need to sample? By what method(s)? Reflect on whether your sampling method will generate a representative sample. Indicate roughly the number of samples and/or observations you intend to collect and your rationale.

[If you would like to save a record of your answer, please type it into this Word document instead of the text box below]

In this investigation there are three combinations of treatment: green compost only, green compost mixed with 25% perlite and green compost mixed with 50% perlite. The statistical populations are each of these three treatments.

Will you need to sample? By what method(s). Reflect on whether your sampling method will generate a representative sample. Indicate roughly the number of samples and/or observations you intend to collect and your rationale.

At this stage in the planning we would not necessarily expect to sample but would examine all plants growing in these growing media.

Step 8 may lead to a revision in the number of samples chosen for the experiment and must always be carried out therefore as part of the planning process.

Check your answer

Back to top

5. Controls

4

Do you need a control(s). If yes - what? If no, why not?

[If you would like to save a record of your answer, please type it into this Word document instead of the text box below]

No. This investigation is a comparison between two fields and there are no other base lines that need to be investigated at the same time.Check your answer

Back to top

6. Variables

5

What is(are) your treatment(s)? What may cause non-treatment variation?

[If you would like to save a record of your answer, please type it into this Word document

What is(are) your treatment(s)?

There is one treatment variable - the effect on growth of adding the perlite to the green compost.

What may cause non-treatment variation?

Variations in the greenhouse environment such as temperature and light

Variation in watering

Grazing by herbivores such as slugs and snails or damage caused by other pests and diseases.

In this investigation the student used multi-trays and we shall consider these in our design. When using multi-trays the edge compartments are likely to experience environmental variation to a greater extent compared to central cells.

Variation between batches of seed and within each batch of seeds in their germination and growth.

Check your answer

6

For each possible cause of non-treatment variation how might the effect be minimised?

[If you would like to save a record of your answer, please type it into this Word document

Potential source of non-treatment variation

How to minimise impact on hypothesis testing.

Variations in the greenhouse such as temperature an light

There are many ways the impact of environmental variation in a greenhouse can be minimised in an experiment. In our design we will seek to randomised our treatments in space and replicate our treatments. If we intend using a multi-tray for convenience then randomising treatments has to be achieved within the constraints of a multi-tray. Again there are several ways this can be achieved, for example you may use blocks laid out in a Latin square. In our design instead we will arrange our three different growing media in order, A in the first cell, B in the second cell, C in the third cell etc. On the second row we will then begin with B, then C then A etc. For the third row we will begin with C, then A then B etc. This stratified approach within a constrained area such as multi-tray should on average minimise the effects of environmental variation. To enhance this further the trays could be turned every few days.

Variation in watering

Solutions to watering are largely dependent on the species of plant you are using. One way forward is to always provide the same amount of water to each cell in each tray. This amount may need to be increased as the plants grow or if the greenhouse temperature increases.

Grazing by herbivores such as slugs and snails or damage caused by other pests and diseases.

Our approach would be to minimise the likelihood that such damage may occur by good greenhouse management. In addition you could build in a degree of redundancy in your experiment by growing more plants than you intend to examine. However you must only reject plants which have clear evidence of damage by pests or diseases. You cannot use this as an excuse to leave out the weak small specimens!

In this investigation the student used multi-trays and we shall consider these in our design. When using multi-trays the edge compartments are likely to experience environmental variation to a greater extent compared to central cells.

A common practice is to consider all the edge plants as a 'guard row' which should act to reduce the effects of environmental variation. These guard row plants are not then evaluated at the end of the experiment.

Variation between the seeds in their germination and growth.

Firstly a single batch of seeds from a known supplier should be used for all treatments. In our design we would then ensure that there are sufficient numbers of replicates for each treatment so that this background variation (sampling error) can be quantified and considered in the analysis of the data.



Check your answer

Back to top

7. Replication

7

Will you use replicates? Why? Indicate how many and explain why this number.

[If you would like to save a record of your answer, please type it into this Word document instead of the text box below]

Yes. To deal with some of the points raised in relation to non-treatment variables considerably more plants may be included in the original design. Having excluded the seeds that did not germinate (a feature that may be evaluated separately), the guard rows and any plants that show clear signs of damage from pests and diseases, a number of replicates may then be sampled from the remaining plants using a random number table. Since this experiment is only one of three carried out by the student he decided to examine 12 plants from each treatment i.e. 12 replicates.

Step 8 may lead to a revision in the number of replicates chosen for the experiment and must always be carried out, therefore, as part of the planning process.

Check your answer

Back to top

8. Statistics

8

Have a go at choosing what might be the correct test to analyse the data from your experiment. Explain your choice.

This is invariably the step that students find the hardest we therefore return to this in interactive exercises in chapters 4 - 8.

[If you would like to save a record of your answer, please type it into this Word document instead of the text box below]

In this experiment there is one treatment variable (type of growth medium) and three samples (green compost only, green compost with 25% perlite and green compost with 50% perlite). The student recorded a number of different parameters for each treatment: time to emergence (when the cotyledons were distinguishable above the soil surface), percentage emergence, fresh and dry weight of above ground material (g), the number of true leaves and leaf area. Using the information in Chapter 4 and appendix b we can make some decisions about which are most likely to be appropriate statistical tests to use.

B.1. What type of investigation am I designing?

In this investigation we are starting out with a question so will be testing a hypothesis.

B.2. Which type of hypotheses am I testing?

There are three types of hypotheses which you need to choose between. If you are not sure which type of hypotheses you will be testing read the information in B.2.1 - B.2.3 of the book before deciding. For more information about hypotheses and hypothesis testing read Chapter 4.

In this example the student wishes to compare samples, he does not have an 'expectation' nor does he wish to test for an association between two or more variables. Therefore the general type of hypothesis for all his observations is:

Do samples come from the same or different populations?

B.2.3. Do samples come from the same or different populations?

There are many tests that will test this type of hypotheses. These tests fall into parametric tests (Chapter 7) to be used when you have Normally distributed data, and non-parametric tests (Chapter 8) when your data are not Normally distributed or you do not know the underlying distribution.

The next step then is to determine if the data are Normally distributed (see Box 3.2 in the book and in the Statistical Software section of the Online Resource Centre). There are 5 criteria to help you tell if your data are likely to be Normally distributed. Only the first can be applied at this planning stage. The remaining criteria are used to confirm your decision when you have actual data.

The first criterion is: a. Are the data measured on an interval scale and are therefore quantitative and continuous such as mm and grams?

With these various measures the answers are:

Data measured

Quantitative and continuous?

time to emergence

Yes

percentage emergence

No

fresh weight

Yes

dry weight

Yes

number of true leaves

No

leaf area

Yes


First let us consider those measures where the data may be parametric, though this will need confirming later when the data are available. If these parametric measures are found to produce non-parametric data then the tests identified for the non-parametric data may be used or the data may be transformed to Normalise it.

B.2.3.1. Parametric tests

From the table it appears that these data may be analyzed using a one-way parametric ANOVA and Tukey's test (7.5 and 7.6)

Experimental design

Test

You have one treatment variable. You are going to compare two samples. The data is unmatched.

t or z test for unmatched data (7.1 or 7.2).

You have one treatment variable. You are going to compare two samples. The data is matched.

t or z test for matched data (7.3)

You have one treatment variable. You are going to compare two or more samples. You wish to test general and specific hypotheses.

One-way parametric ANOVA and Tukey's test (7.5 and 7.6)

You have two treatment variables. Each variable has at least two categories or classes and all categories from one variable are combined with all categories from the second variable. You wish to test general and specific hypotheses.

Two-way parametric ANOVA and Tukey's test (7.7. and 7.8.)

You have two treatment variables. Each variable has at least two categories. One variable is randomised or nested with respect to the second variable. You wish to test general hypotheses.

Two-way nested ANOVA (7.9.)

You have three treatment variables. Each variable has at least two categories and all categories from each variable are combined with all other categories from the other variables. You wish to test general and specific hypotheses.

Three-way parametric ANOVA (7.10.)

None of the above

Chapter 8. and Sokal & Rohlf, 1981.



If you wish to use the one-way parametric ANOVA the following criteria need to be met. You:

  1. Wish to test for differences in population means.
  2. Have one treatment variable.
  3. Have parametric data.
  4. Have an experimental design, which means that each item is assigned at random to the samples.
  5. Have samples where the variation is similar (homogeneous).
  6. Have the same number of replicates (observations) in each sample.

In our planning stages we can confirm that all criteria other than 5 have been met with our current design and criterion 5 will need to be checked when the data have been collected.

B.2.3.2. Non-parametric tests

In our design there are two measurements which we know are non-parametric scales (percentage emergence and number of true leaves). For the latter (number of true leaves) since the experimental design is the same as that for the parametric measures it is not surprising that the most appropriate analysis is the non-parametric equivalent i.e. either the non-parametric one-way ANOVA or the Scheirer - Ray - Hare test. Given the amount of data we recommend the latter.

Experimental design

Test

You have one treatment variable. You are going to compare two samples. The data is unmatched. You have 20 observations or less in each sample.

Mann Whitney U test (8.1.)

You have one treatment variable. You are going to compare two samples. The data is unmatched. The data is measured on a continuous scale and you have more than 30 observations in each sample.

z test for unmatched data (7.1.)

You have one treatment variable. You are going to compare two samples. The data is unmatched. You have more than 20 observations in each sample.

Sokal & Rohlf, 1981.

You have one treatment variable. You are going to compare two samples. The data is matched. You have less than 30 pairs of observations.

Wilcoxon's rank paired test (8.2.)

You have one treatment variable. You are going to compare two samples. The data is matched. You have more than 30 pairs of observations.

z test for matched data (Chapter 7 (7.2)).

You have one treatment variable. You are going to compare two or more samples. You wish to test general and specific hypotheses.

One-way ANOVA (Kruskal Wallis test)( 8.3. and 8.4)

You have more than one treatment variable. You are going to compare two or more samples. You wish to test general and specific hypotheses. You will be using a calculator.

Two-way non parametric ANOVA (8.5. and 8.6)

You have more than one treatment variable. You are going to compare two or more samples. You wish to test general hypotheses. You want to use a computer.

Scheirer - Ray - Hare test (8.7.).



To use the Scheirer-Ray-Hare test you:

  1. Wish to test for differences in population medians.
  2. Have two treatment variables each with at least two categories.
  3. The design is orthogonal.
  4. Have non-parametric data that can be ranked.

At this stage in our planning we can confirm that all these criteria are met. Therefore, unless our design changes we should be able to use this test to examine the hypotheses relating to the number of true leaves.

For the percentage emergence we will have three samples with data on the number of seeds emerging in growing medium A and the number of seeds not emerging in growing medium A (with the same for the media B and C). This data is most usefully analyzed using a chi-squared or G tests for association.

To use the chi-squared or G tests for association (5.3.1.) you:

  1. Wish to test for an association between two treatment variables.
  2. Have data that is organised into more than two categories for at least one of the variables and into two or more categories for the second variable.
  3. Have data that is counts or frequencies and is not percentages or proportions.
  4. Have observations that are independent of each other.
  5. Have expected values that are more than 5.

At this stage in our planning we can confirm that all the criteria other than criterion 5 are met. This final criterion can only be checked when the data are collected. However in the current design each type of compost should be present in 34 cells excluding the guard rows and therefore it is likely that this criterion will be met.

Check your answer

9

Finalise your aim and objectives.

[If you would like to save a record of your answer, please type it into this Word document instead of the text box below]

In this example we only examined one of the three original objectives. We therefore only include this one objective here.

Aim:

Assessing the effects of perlite and humic acid as ameliorants for green compost when growing Ocimum basilicum (sweet basil).

Objective:

The effect of perlite as a bulking agent (no perlite, 25% perlite, 50% perlite) in green compost on the growth of Ocimum basilicum as indicted by time to emergence, percentage emergence, dry and fresh weight (g), number of true leaves and leaf area after 4 weeks.

Check your answer

10

Try drafting your hypotheses.

[If you would like to save a record of your answer, please type it into this Word document instead of the text box below]

Parametric and non-parametric ANOVAs and the Sheirer-Ray-Hare test both test similar hypotheses. The parametric ANOVA tests population means, and the non-parametric ANOVA compares population medians. The hypotheses will be the same for each measure such as time to emergence etc. We illustrate all the hypotheses for these tests in relation to the parametric measure 'time to emergence'.

H0: There is no difference between the mean time to emergence (hours) of Ocimum basilicum between the three growing media.

H1: There is a difference between the mean time to emergence (hours) of Ocimum basilicum between the three growing media.

For the chi-squared or G tests for association the hypotheses would be:

H0: There is no association between the number of plants of Ocimum basilicum emerging in the three growing media.

H1: There is an association between the number of plants of Ocimum basilicum emerging in the three growing media.

Check your answer

Back to top

9. Influencing outcomes

11

Is this relevant to your experiment? In what way?

[If you would like to save a record of your answer, please type it into this Word document instead of the text box below]

It is unlikely that the investigator could influence the outcomes of this investigation.Check your answer

Back to top

10. Assumptions and Bias

12

List all the assumptions and possible causes of bias in your design.

[If you would like to save a record of your answer, please type it into this Word document instead of the text box below]

The non-treatment effects are minimised and do not unduly affect the data.

The sampling of plants does produce a representative sample for each treatment.

The batch of seeds being used is representative of the seed supplied generally for Ocimum basilicum.

Check your answer

Back to top

11. Repeatability

13

Could this experiment be repeated?

[If you would like to save a record of your answer, please type it into this Word document instead of the text box below]

a) Yes
b) No
That right!Incorrect. It is possible to repeat this experiment.
Check your answer

Back to top

12. Back to the beginning

14

Provide a complete review of your experimental design written in such a way that another person could use this review to carry out your experiment.

[If you would like to save a record of your answer, please type it into this Word document instead of the text box below]

Aim: Assessing the effects of perlite and humic acid as ameliorants for green compost when growing Ocimum basilicum (sweet basil).

Objective:

The effect of perlite as a bulking agent (no perlite, 25% perlite, 50% perlite) in green compost on the growth of Ocimum basilicum as indicted by time to emergence, percentage emergence, dry and fresh weight (g), number of true leaves and leaf area after 4 weeks.

Experimental design:

The following is a summary of our experimental design. Clearly there are some further details that need to be included before this is a finalised method.

A batch of Ocimum basilicum seedwas obtained from a known supplier. A green compost identified in an earlier experiment as a suitable growing medium for this species was mixed with varying amounts of an inert bulking agent (perlite). The growth media mixes were: no perlite (A), 25% perlite (B), 50% perlite (C). A multi-tray (15 x 10 cells) containing a total of 150 cells was filled with the various growing media. The first cell in the first row was filled with medium A, the second cell in this row was filled with medium B and the third was filled with medium C. This was repeated until the first row was completed. In the second row, medium B was used first followed by C and then A etc. The third and subsequent rows were similarly altered. All outer rows on the multi tray were guard rows and not used for subsequent evaluation. The multi-tray was watered to saturation and a 5mm depression made in the growing medium in the centre of each cell. One seed was placed in each and gently covered over. The multi-tray was placed in a greenhouse and the plants left to germinate and grow for 4 weeks. The time to emergence was recorded for each seed and percentage emergence established for each growing medium excluding the guard rows. After 4 weeks the plants were examined and where clear evidence of damage by pest or disease was identified these plants were no longer included in the experiment. A random number table was used to identify a sample of 12 plants for each growing medium and these were cut across at the base. Each plant was then examined and the following recorded: dry and fresh weight (g), number of true leaves and total leaf area.

It is anticipated that at least two of these measures (percentage emergence and number of true leaves) will generate non-parametric data. For the first a chi-squared or G test for association would be appropriate. The number of true leaves could be analyzed using a Scheirer-Ray-Hare test. Where data are confirmed as being parametric then a one-way parametric ANOVA and Tukeys test should be considered. The hypotheses to be tested will depend on the test being used. For example most of this data would be tested by an ANOVA or the Scheirer-Ray-Hare adaptation of the ANOVA. These hypotheses can be illustrated in terms of the mean time to emergence.

H0: There is no difference between the mean time to emergence (hours) of Ocimum basilicum between the three growing media.

H1: There is a difference between the mean time to emergence (hours) of Ocimum basilicum between the three growing media.

Where the chi-squared or G test for association is used the hypotheses would be:

H0: There is no association between the number of plants of Ocimum basilicum emerging in the three growing media.

H1: There is an association between the number of plants of Ocimum basilicum emerging in the three growing media.

Check your answer

Back to top

Reflection

15

Describe how you felt at the various stages of planning this experiment. Can you identify one feature of your experience that you can seek to improve? How does our answer differ from yours? In what ways were your ideas better?

Well done! You have now completed this interactive exercise. You may now wish to print this page for your reference, and also the Word documents if you wish to keep a permanent record of your answers.Check your answer