Resources
In an undergraduate plant physiology laboratory, students test the success of vegetative propagation via adventitious rooting of stem cuttings in Coleus scutellarioides treated with exogenous auxin (indole-3-butyric acid, IBA) compared to sexual propagation via seed germination. Tray 1 contains stem cuttings treated with IBA powder. Tray 2 contains sexually produced seeds planted in sterile vermiculite. After , out of cuttings have successfully formed adventitious root systems long. Among the seeds, out of have germinated and developed radicles. (a) Calculate the percentage success rate for both vegetative propagation and sexual seed germination. (b) Formulate the null hypothesis () for a chi-square () test of independence comparing the propagation success rates between these two methods. (c) Given the observed contingency data, compute the expected count of successful propagules for the cutting treatment under .
Answer and steps
Final answer
(a) Vegetative cuttings: ; Seed germination: ; (b) : The probability of propagation success is independent of the reproductive mode (vegetative cutting vs. seed germination); (c) Total successful ; total trials ; Expected count for cuttings .
Steps
Part (a): Calculating the Percentage Success Rates
To find the success rate for each propagation method, divide the number of successful outcomes by the total sample size () and multiply by :
- Vegetative propagation (stem cuttings treated with IBA):
- Sexual propagation (seed germination):
Part (b): Formulating the Null Hypothesis ()
A chi-square () test of independence evaluates whether two categorical variables—here, propagation method (asexual cutting vs. sexual seed) and outcome (success vs. failure)—are statistically independent.
- Null hypothesis (): The probability of propagation success is independent of the reproductive mode (vegetative cutting vs. seed germination). That is, there is no association between the method of propagation and whether a propagule successfully establishes.
Part (c): Computing the Expected Frequency Under
To find the expected count for a cell in a contingency table, construct the table of observed counts:
Under the null hypothesis of independence, the expected frequency () for any cell is calculated using the marginal totals:
For the cell corresponding to successful cuttings:
Common Pitfall: Do not confuse the observed count () with the expected count (). The expected count represents how many cuttings would succeed purely by chance if both methods shared the pooled overall success rate of .
Typical errors
01 The student assumes equal distribution across all cells or treatments under the null hypothesis instead of weighting by the marginal success total, resulting in an expected count of 25.
Sample Student Work
Reviewed by ACEpath TeacherStep-by-Step Review
Walk Through It with Your Teacher- Step 1
- Step 2
- Step 3Correct
In a chi-square test of independence, the null hypothesis states that the outcome is independent of the group, not that success and failure have equal probability. You need to use the pooled overall success rate across both treatments, which gives an expected count of .
WHAT YOU DID WELL
Compliments from Your TeacherYou correctly computed both success percentages from the experimental counts.
You clearly articulated an appropriate null hypothesis for the two-sample comparison.
THE ERROR
Where Your ACEpath Teacher Spotted the Mistakeno
WHY IT HAPPENED
ACEpath Teacher's DiagnosisYou treated the null hypothesis of independence as asserting an equal 1:1 ratio between success and failure (equiprobability), rather than calculating the expected count from the marginal row and column totals ().
REASONING PATH
Your ACEpath teacher’s suggested approach- Independence vs. Equal Outcomes
In a test of independence, means the outcome distribution is the same across groups, not that each outcome is equally likely ().
Always find the pooled overall proportion across all groups first before calculating cell expectations.
- Expected Cell Frequency Formula
, where is the row total, is the column total, and is the grand total.
Construct a contingency table with row and column sums, then multiply the relevant row and column margins and divide by the grand total.
CORRECTED PATH
ACEpath Teacher's DemonstrationConstruct the contingency table: Cuttings row total is , and total successful propagules across both treatments is .
Identify the grand total number of trials across both treatments: .
Apply the expected frequency formula: .
Expected count for successful cuttings
02 The student confuses the null hypothesis of independence with a directional alternative hypothesis stating that cuttings will outperform seeds.
Sample Student Work
Reviewed by ACEpath TeacherStep-by-Step Review
Walk Through It with Your Teacher- Step 1
- Step 2
Stem cuttings treated with IBA have a significantly higher success rate than seeds.
Correct: Propagation success is independent of the propagation method (or there is no difference in the success rates between vegetative cuttings and seeds).
You wrote a directional alternative hypothesis () predicting an effect. A null hypothesis () must state the baseline condition of 'no effect', 'no difference', or 'independence' between the two categorical variables.
- Step 3
WHAT YOU DID WELL
Compliments from Your TeacherYou calculated both empirical success percentages accurately from the observed data.
You correctly identified the row, column, and grand totals to compute the expected frequency under independence.
THE ERROR
Where Your ACEpath Teacher Spotted the Mistakeno
WHY IT HAPPENED
ACEpath Teacher's DiagnosisYou confused the null hypothesis (), which posits no association or difference, with a directional research or alternative hypothesis () that posits an experimental advantage.
REASONING PATH
Your ACEpath teacher’s suggested approach- Defining the Null Hypothesis ($H_0$)
In hypothesis testing, represents the default assumption of no effect, equality, or independence between variables.
State in terms of 'no difference' or 'independence' rather than predicting which group will perform better.
- Distinguishing $H_0$ from $H_a$
The alternative hypothesis ( or ) asserts an effect or relationship (e.g., group A > group B), whereas asserts that observed differences are due solely to chance.
Reserve statements of inequality or superiority (such as 'significantly higher') exclusively for the alternative hypothesis.
CORRECTED PATH
ACEpath Teacher's DemonstrationRecall that a chi-square test of independence evaluates whether two categorical factors (propagation method and rooting success) are unrelated.
Formulate to express that no association exists: 'The probability of rooting success is independent of the propagation method (or the success rate is the same for cuttings and seeds).'
(a) Vegetative cuttings: , Seed germination: ; (b) : Propagation success is independent of the reproductive mode (vegetative cutting vs. seed germination); (c) Expected count
03 The student calculates the expected count of failures instead of successes for the cutting treatment, yielding 11.5.
Sample Student Work
Reviewed by ACEpath TeacherStep-by-Step Review
Walk Through It with Your Teacher- Step 1
$ = = 92% = = 62%$$
- Step 2
- Step 3Correct
You applied the expected value formula using the column total for failures () instead of the column total for successes (), giving the expected count of failed cuttings rather than successful cuttings.
WHAT YOU DID WELL
Compliments from Your TeacherYou correctly computed the success percentages for both propagation treatments.
You accurately formulated the null hypothesis of independence between propagation method and outcome.
You correctly recalled and applied the contingency table expected-frequency formula .
THE ERROR
Where Your ACEpath Teacher Spotted the MistakeYou selected the column total of failures instead of successes when computing the expected count for successful cuttings.
WHY IT HAPPENED
ACEpath Teacher's DiagnosisYou correctly knew the formula for expected counts in a contingency table, but you mapped the target cell to 'failures' rather than 'successes' before calculating the column marginal total.
REASONING PATH
Your ACEpath teacher’s suggested approach- Align Target Cell with Marginal Totals
For any contingency table cell, , where the column total must match the exact category (e.g., success vs. failure) requested.
Check whether the problem asks for successful or failed propagules before summing the column counts.
CORRECTED PATH
ACEpath Teacher's DemonstrationIdentify the target cell: row = cuttings (), column = successful propagules.
Calculate the column total for successful propagules: .
Compute the expected count: .
(a) Vegetative cuttings: , Seed germination: ; (b) : Propagation success is independent of the propagation method; (c)
Lock it in
In a chi-square test of independence for a contingency table, what does the null hypothesis () assume about the two variables?
It assumes the probability of an outcome is independent of group membership; each group shares the same pooled outcome proportion.
Similar question
A botanist tests rooting success in willow cuttings and boxwood cuttings. Overall, cuttings root successfully while fail. What is the expected count of successfully rooted boxwood cuttings under the null hypothesis of independence?
Answer
. The row total for boxwood is , the column total for success is , and the grand total is .
In a seed stratification experiment, cold-treated seeds produce seedlings, while untreated control seeds produce seedlings. Under for a chi-square test of independence, what is the expected count of germination failures in the cold-treated group?
Answer
. The cold-treated row total is . The total number of failures is . The grand total is . Therefore, .
A researcher assesses grafting success across three scion cultivars (A, B, and C) with sample sizes of , , and grafts respectively. Across all grafts, successfully establish. Find the expected number of successful grafts for Cultivar A and determine whether assuming equal distribution across all six cells would give a correct result.
Answer
. The pooled success rate is , so for Cultivar A (), . Assuming equal distribution across all 6 cells would yield , which is incorrect because it ignores both the overall success rate and the unequal cultivar sample sizes.