Paper tutorial / Research in practice

Sustainability, investment and profitability with fsQCA

Use QuickPLS fsQCA to find combinations of sustainability practices associated with a profitability assessment in fictional businesses.

18 min read

We use this paper to show how a similar analysis can be carried out in QuickPLS. Follow the steps with the supplied practice data, then adapt the setup and calculations to your own study. The practice data are a separate teaching example.

Why can similar sustainability scores accompany different financial outcomes?

A high sustainability score may describe a business with strong research spending, or one with efficient operations and limited research. A configuration study asks which combinations accompany profitability. It does not assume that raising one score has the same association in every business.

Yeung, Chen and Tang used fsQCA to study 1,380 non-financial firms with environmental disclosure ratings of at least B. Their nine conditions combined environmental, social and governance measures with financial and contextual factors, and their outcome was return on assets. The lesson below uses a smaller, original five-condition teaching model. Its ratings are not financial accounts, ESG ratings or a reproduction of the paper.

Compare five ingredients of a fictional investment profile

Use 220 fictional established businesses. Each row contains an environmental operating score, a workforce score, a governance score, an innovation investment score, an asset renewal score and a profitability assessment. Ask whether different combinations of these five ingredients are sufficient for strong profitability assessment.

The five-condition design has 32 possible truth-table configurations. This reduction from the paper's nine conditions is deliberate: it lets you inspect the relationship between case support and the number of possible configurations. It does not show that the omitted financial or institutional conditions are unimportant.

PROFIT is an invented assessment on the same continuous 1–7 range as the conditions. Do not label it ROA, a percentage return or a monetary outcome. A real financial study would require an accounting definition, reporting period, currency treatment where relevant and a defensible calibration for negative and positive returns.

Conditions and outcomes in this practice study. The named conditions enter configuration analysis together. Each outcome is a separate question; the arrow is not a regression effect.
The named conditions enter configuration analysis together. Each outcome is a separate question; the arrow is not a regression effect.

On a small screen, swipe the table sideways to see every column.

ColumnMeaningAnalytical role
ENVEnvironmental operating practicesCondition
WORKWorkforce practicesCondition
GOVGovernance oversightCondition
INNOVInnovation investment intensityCondition
RENEWAsset renewal activityCondition
PROFITStrong profitability assessmentOutcome
case_idFictional businessLabel

Step 1. Separate financial meaning from convenient scaling

In a writable project choose File > Import Data…, then Choose data file…. Select the CSV and inspect Review & import. Confirm raw data, Header row 1, comma separator and Decimal convention Dot (1.25), with the analytical numeric columns set to Continuous. Choose Import dataset and confirm Dataset imported and verified.

Import the complete CSV and check that each assessment is numeric and lies within the stated 1–7 scale. Inspect low as well as high values. Leave case_id out of the analytical variables. There are no reverse-coded items and no supplied missing cases.

For this classroom rubric, scores around 2 indicate weak evidence of the attribute, 4 indicate a borderline assessment and 6 indicate strong evidence. Use those values as exclusion, crossover and inclusion anchors. The anchors describe this invented rubric; they cannot be copied into an analysis of reported percentages.

Be particularly careful with innovation spending. Zero spending is not the same as a missing disclosure. An industry percentile can express relative standing within an industry, but it changes the meaning of the set from an absolute investment level. Decide whether relative or absolute standing answers your actual question before selecting a calibration.

Screenshot to be addedImport the 220-case teaching file

Step 2. Calculate the PROFIT baseline in QuickPLS

A completed calculation is the beginning of interpretation. First confirm the dataset, case count, calibration and thresholds shown with the result. Then read the expression and its fit. Otherwise you can accidentally compare an edited setup with a result calculated from earlier choices.

  1. Open the fsQCA settings

    With the imported dataset selected, choose Calculate…. In Find a method select Fuzzy-set Qualitative Comparative Analysis. Open Method settings. The form starts with Conditions and Outcome and case identity.

  2. Assign the conditions and outcome

    Under Conditions choose Add condition for each of ENV, WORK, GOV, INNOV and RENEW, in that order. Under Outcome and case identity select PROFIT as Outcome variable, Presence of the outcome as Analyse and case_id as Case identifiers. Check that only the intended conditions appear; the case label is not an analytical condition.

  3. Calibrate every selected column

    For each condition and the outcome choose Direct fuzzy-set calibration. Enter exclusion 2, crossover 4 and inclusion 6. Explain the column’s set definition and the invented assessment rubric in its anchor rationale. These increasing anchors make a higher raw score correspond to greater membership. There is no separate inclusion-membership entry in this form.

  4. Set the baseline thresholds

    Under Truth table and solution assumptions set Frequency threshold to 2, Consistency threshold to 0.80 and PRI threshold to 0.50. Leave Request intermediate solutions with explicit directional expectations unchecked for this baseline. The reference answer uses the conservative family from the completed output; it is not a family selected in this settings form.

  5. Calculate and check the result

    Choose Start calculation and wait for completion. Check the outcome, calibration, thresholds and included-case count in the completed result before interpreting it. The reference includes all 220 cases. If your result differs, inspect Exclusions and the selected settings rather than changing the answer to fit the reference.

  6. Read one complete solution

    In Results, select the completed calculation in Result. Under Estimates open Truth table and read frequency, consistency and PRI together. Then open Overall solution fit — every alternative. Use the conservative rows to compare with this lesson’s reference answers; keep the Solution identifier with the family. In Pathway fit and coverage — every alternative, read all pathways belonging to that same family and solution. Do not combine pathways from different complete alternatives. An alternative number identifies an expression; it does not rank its quality.

  7. Check the case memberships

    Under Model and data open Case memberships and source rows. Locate C0014 by its case identifier and compare INNOV and RENEW with PROFIT. Use the lesson’s checked case table to follow the membership calculation. One case illustrates the pathway; its membership is not the pathway’s overall consistency.

Screenshot to be addedSelect the PROFIT model in Method settings
Screenshot to be addedSet the truth-table thresholds

Step 3. Check the answers for the 220-case file

Use direct logistic calibration for every selected condition and outcome: exclusion 2, crossover 4, inclusion 6; the calibration used by the reference assigns membership 0.95 at the inclusion anchor. The baseline uses frequency 2, row consistency 0.80 and PRI 0.50. Read the conservative rows in the completed solution tables. 220 of 220 cases are included; excluded case identities, if any, are in the download.

For PROFIT, 11 truth-table rows are positive, 21 negative and 0 remainders. There are 1 complete conservative alternative(s). Alternative 1 has consistency 0.831, coverage 0.699 and PRI 0.663.

About these answers: Reference answers calculated with R QCA and independently checked. These tables are not outputs from a QuickPLS app run.

On a small screen, swipe the table sideways to see every column.

OutcomeSelected alternativeSolution consistencySolution coverage
PROFIT1 of 10.8310.699
Screenshot to be addedRead the completed PROFIT result
What the separate checks cover

1,644 numerical comparisons passed tolerance 0.000000000001. Python independently recalculated memberships, case assignments, supported-row consistency and PRI, pathway and solution fit, and checked that each conservative alternative covers exactly the positive Boolean rows. This does not prove that the minimiser found every globally minimum expression and does not execute QuickPLS.

Step 4. Read the complete PROFIT alternative

The table gives every pathway in reference alternative 1. Read * as AND, ~ as absence, and the rows together as OR. An omitted condition can be present or absent within that term. Keep the complete expression together. The downloadable calculation files include any other returned alternatives.

The PROFIT outcome is a calibrated membership, not a pass/fail prediction. Raw coverages can overlap; adding them does not give solution coverage.

About these answers: Reference answer from R QCA 3.25, with independent checks of calibration, pathways and fit.

On a small screen, swipe the table sideways to see every column.

PathwayConsistencyRaw coverageUnique coverage
INNOV*RENEW0.8570.5870.194
ENV*GOV*INNOV0.8800.4630.016
ENV*~WORK*GOV*~RENEW0.8810.3500.041

Step 5. Follow one actual practice case: C0014

For case C0014, the membership in INNOV*RENEW is 0.521. Take the smallest membership across the conditions in that term, complementing a membership first when it has ~. The outcome membership is 0.898. This is one case inside a pathway, not the pathway’s overall consistency.

About these answers: Values calculated from this CSV using the stated logistic calibration.

On a small screen, swipe the table sideways to see every column.

Condition or outcomeRaw scoreCalibrated membership
ENV3.52220.331
WORK4.56040.695
GOV4.06510.524
INNOV4.05580.521
RENEW4.77520.758
PROFIT5.47540.898
Screenshot to be addedFind case C0014 in the result

Step 6. Compare the planned threshold changes

The frequency / consistency column lists the rules applied to truth-table rows. The final consistency column measures the minimised solution. These are different quantities; a row threshold of 0.85 does not require solution consistency to exceed 0.85. A frequency increase can remove observed support and create more remainders.

These runs change frequency or row consistency while holding calibration fixed. They do not test other calibration anchors, outcome negation or another solution family. If the lesson asks you to explore those choices, save and report them as additional runs.

About these answers: External reference runs, with every returned conservative alternative retained.

On a small screen, swipe the table sideways to see every column.

Outcome / changeFrequency / consistencyPositive rowsAlternativesSolution consistencySolution coverage
PROFIT: baseline2 / 0.801110.8310.699
PROFIT: frequency3 / 0.801110.8310.699
PROFIT: consistency2 / 0.851110.8310.699

Step 7. Distinguish a rare route from a well-supported route

Frequency describes the support for a truth-table row, not the number of times a pathway appears in a report. A minimised pathway can cover several rows, while one populated row may contain several businesses. Move between the row and case views when a pathway seems unexpectedly broad.

A frequency threshold of 2 admits configurations with at least two cases under the truth-table classification. It does not establish a universal minimum sample size. A much larger real dataset may justify a different support rule because the distribution of cases and the research purpose differ.

Necessity and sufficiency ask different containment questions. Strong governance could occur in nearly every highly profitable case but also in many less profitable cases. That pattern could be consistent with necessity without making governance sufficient. Read the relevant necessity fit and relevance information instead of inferring necessity from a prominent solution term.

In Model and data > Case memberships and source rows, locate C0014 and compare INNOV and RENEW with PROFIT. Use the matching case table below to check the minimum operation. A single case helps explain a pathway; it does not supply the pathway’s overall consistency.

On a small screen, swipe the table sideways to see every column.

Table or viewFirst checkInterpretation task
Case memberships and source rowsOriginal score beside membershipDoes the calibration express the intended assessment?
Truth tableFrequency, consistency and PRI togetherWhich configurations have enough and sufficiently clear support?
Necessity outputConsistency and relevance for each conditionIs the outcome contained in the condition, and is that informative?
SolutionComplete alternative and covered casesWhich conjunctions describe outcome membership?
Pathway fitRaw and unique coverageIs this route broad, narrow or largely overlapping?

A broad governance route can overlap an innovation route

Hypothetical example

Illustrative reading example

Consider two invented pathways: ENV * GOV and INNOV * RENEW. The first describes environmental operations together with oversight. The second describes innovation together with asset renewal. Neither says its individual ingredients work alone.

Suppose the first pathway has consistency 0.89 and raw coverage 0.62; the second has consistency 0.92 and raw coverage 0.38. A larger consistency does not make the second pathway more common. It may describe a smaller group with clearer agreement.

Suppose their overlap in outcome membership is 0.24. The solution coverage is then 0.62 + 0.38 − 0.24 = 0.76. The unique coverages are 0.38 and 0.14. You cannot obtain solution consistency by averaging 0.89 and 0.92, because its denominator depends on the union of case memberships.

The two hypothetical pathways cover 76% of the outcome membership after accounting for overlap. This statement concerns set coverage, not explained variance, accounting profit gained or the probability that an investment strategy will succeed. None of these values was fitted to the downloadable data.

Step 8. Test the dependence on sparse support and a stricter outcome

Plan two separate comparisons. First increase frequency from 2 to 3 while retaining the calibration and consistency threshold. Inspect which configurations lose eligibility and which pathways change. Second return to frequency 2 and increase consistency from 0.80 to 0.85. This tests a different judgement about sufficiency.

If strong profitability could reasonably require a more demanding score, save a calibration alternative for PROFIT alone. Explain the new substantive threshold before calculating. Changing the outcome boundary changes the set you are trying to explain, so its coverage cannot be compared as though the outcome stayed identical.

For an optional weak-profitability analysis, reopen the fsQCA calculation settings, keep PROFIT as the outcome and set Analyse to Negation of the outcome (separate analysis). Start a new calculation and read its own result. A strong-profits pathway with innovation present does not imply that innovation’s absence explains weak profits. The supplied answer tables concern outcome presence; they do not answer this additional question.

For real accounts, examine whether results depend on one reporting year, industry composition or unusual asset values. These checks need suitable data and a declared design. The synthetic single-period file cannot establish stability across economic cycles.

Keep the financial and set interpretations straight

These checks protect the meaning of the analysis. They are not instructions to search for the largest coverage or the cleanest managerial recommendation. A qualified answer about a narrow group is more useful than a broad claim unsupported by the cases.

On a small screen, swipe the table sideways to see every column.

SymptomCheckAction
Every route contains INNOVNecessity results and calibrated innovation distributionDistinguish frequent appearance from individual necessity.
Profitability memberships cluster near 0.5Whether the outcome rubric is genuinely discriminatingReview the measurement definition and report ambiguous cases; do not spread memberships artificially.
A rare route has the highest consistencyIts raw and unique coverage and casesDescribe its limited reach rather than naming it the best strategy.
Coverage falls after raising the outcome anchorWhether the outcome set changedReport the revised definition and case memberships beside the comparison.
An apparent financial loss is entered as missingOriginal accounting definition and codingKeep observed losses as values and choose a defensible calibration.
The solution changes when frequency risesWhich configurations fell below the support ruleReport sensitivity to sparse support instead of removing the affected cases.

Step 9. Write a configuration result, not an investment forecast

For Results, write: The selected [family] solution contained [number] pathways. The pathway [expression] showed consistency [value], raw coverage [value] and unique coverage [value], with [cases] illustrating its scope. Under [alternative], [specific change] occurred. Do not describe the largest pathway as a proven investment recommendation.

Report itemRequired detail
Outcome definitionAssessment rubric or actual accounting measure and period
CalibrationAnchors, direction and reasons
Case supportIncluded cases and truth-table frequency rule
SolutionsWhole alternatives, pathway fit and solution fit
SensitivityChanged choices and affected configurations
LimitsAssociational configurations, sample and period boundaries

Methods writing outline

We used [included count] fictional business profiles to practise a five-condition fsQCA analysis. Assessment scores were calibrated using [anchors and rationale]. The baseline used frequency [value], consistency [value] and PRI [value]. We examined necessity and sufficiency separately and compared [named alternatives]. The teaching ratings are not accounting or ESG observations.

Step 10. Write the result for this teaching file

We calibrated 220 synthetic cases using direct logistic calibration with anchors 2, 4 and 6 and inclusion membership 0.95. We used a frequency requirement of 2, row consistency of 0.80 and PRI of 0.50. We retained the conservative solution family.

Report the complete selected alternative and the changes shown in the sensitivity table. Explain case examples alongside aggregate coverage. These results describe the original teaching dataset, not the source paper’s observations. If several alternatives exist, explain the selection and preserve the others; do not combine favourite terms.

Check that the conclusion matches the evidence

  • Explain why PROFIT is not ROA.
  • Calculate overlap and unique coverage from the worked example.
  • Name a difference between frequency and coverage.
  • Explain why a high-consistency route may describe few cases.
  • Keep outcome recalibration separate from a support-threshold change.
  • Report the weak-profitability analysis from its own calculation.

Further reading

Related guides