Choice-Based Conjoint (CBC) Analysis
Analyse discrete choice experiments in which respondents pick one alternative from each choice set. A conditional (multinomial) logit model estimates part-worths, attribute importance and willingness to pay when price is included.
Updated · By Bibin XavierData 3600 × 8
| Respondenttext · 150 levels | Tasknumeric | Alternativenumeric · coded | Chosennumeric · coded | Brandtext · 3 levels | Pricenumeric · coded | Storagetext · 3 levels | Warrantytext · 2 levels |
|---|---|---|---|---|---|---|---|
| R001 | 1 | 1 | 1 | Apex | 299 | 64 GB | 1 year |
| R001 | 1 | 2 | 0 | Zenit | 399 | 128 GB | 1 year |
| R001 | 1 | 3 | 0 | Apex | 499 | 128 GB | 1 year |
| R001 | 2 | 1 | 0 | Nova | 499 | 64 GB | 1 year |
| R001 | 2 | 2 | 0 | Zenit | 299 | 64 GB | 2 years |
| R001 | 2 | 3 | 1 | Nova | 399 | 64 GB | 2 years |
| R001 | 3 | 1 | 0 | Apex | 299 | 64 GB | 1 year |
| R001 | 3 | 2 | 1 | Nova | 299 | 128 GB | 2 years |
Variables and options
Results
3600 rows · example dataConditional logit coefficients
| Parameter | Coefficient | SE | z | p | 95% CI |
|---|---|---|---|---|---|
| Brand: Nova | 0.5442 | 0.0989 | 5.505 | < .001 | [0.3505, 0.7380] |
| Brand: Zenit | 0.2857 | 0.0996 | 2.867 | .004 | [0.0904, 0.4810] |
| Storage: 128 GB | 0.6763 | 0.1016 | 6.654 | < .001 | [0.4771, 0.8756] |
| Storage: 256 GB | 0.9169 | 0.1030 | 8.905 | < .001 | [0.7151, 1.1187] |
| Warranty: 2 years | 0.4809 | 0.0799 | 6.016 | < .001 | [0.3243, 0.6376] |
| Price (per unit) | -0.0094 | 0.0006 | -16.886 | < .001 | [-0.0105, -0.0083] |
Note. Dummy coded against the first level of each attribute (Brand: Apex; Storage: 64 GB; Warranty: 1 year).
Part-worths and importance
| Attribute | Level | Part-worth | Willingness to pay | Importance |
|---|---|---|---|---|
| Brand | Apex | -0.277 | 0.00 | 14.2% |
| Nova | 0.268 | 57.79 | ||
| Zenit | 0.009 | 30.34 | ||
| Storage | 64 GB | -0.531 | 0.00 | 24.0% |
| 128 GB | 0.145 | 71.82 | ||
| 256 GB | 0.386 | 97.37 | ||
| Warranty | 1 year | -0.240 | 0.00 | 12.6% |
| 2 years | 0.240 | 51.08 | ||
| Price | linear | -0.0094 | — | 49.2% |
Note. Willingness to pay for each level relative to the first level = −βlevel / βprice, in the units of Price.
Attribute importance
Interpretation
Price has the largest influence on choice (49.2%). One unit of utility is worth 106.20 in Price units; the willingness-to-pay column converts each level into that currency.
How to cite this tool
Using these results in a thesis, journal article, report or class assignment? Please cite this free tool so others can find it. It keeps the site free.
APA 7
Xavier, B. (2026). Choice-Based Conjoint (CBC) Analysis [Web application]. FreeStatisticalTools.com. https://freestatisticaltools.com/choice-based-conjoint-analysis/
BibTeX
@software{xavier_choicebasedconjoint_2026,
author = {Xavier, Bibin},
title = {Choice-Based Conjoint (CBC) Analysis},
publisher = {FreeStatisticalTools.com},
year = {2026},
url = {https://freestatisticaltools.com/choice-based-conjoint-analysis/},
note = {Web application}
}Suggested sentence for your methods section: “The analysis was carried out with the free Choice-Based Conjoint (CBC) Analysis tool on FreeStatisticalTools.com (Xavier, 2026).”
What is choice-based conjoint?
In a choice-based conjoint study, respondents choose their preferred alternative from several sets of product options. A conditional logit model estimates how each attribute level affects the probability of choice, from which part-worths, importances and willingness to pay follow.
When to use choice-based conjoint
Use CBC when respondents choose between alternatives in repeated choice sets, the most realistic and now most common form of conjoint analysis.
Assumptions
- Independence of irrelevant alternatives (conditional logit).
- Exactly one chosen alternative per choice set.
- Preferences are homogeneous across respondents in this aggregate model; hierarchical Bayes or mixed logit relaxes this.
How to set up your data
Long format: one row per alternative in each choice set, with respondent and task identifiers, a 0/1 choice column and the attribute levels.
How to use this tool
- Enter your data. Paste columns from Excel, Google Sheets or SPSS with variable names in the first row, or upload a CSV or Excel file. The example data shows the expected layout.
- Choose the variables. Select the choice indicator, value meaning chosen, respondent identifier and choice task identifier.
- Read the results. Results update automatically. Read the plain-English interpretation, then use Copy table to paste APA-style tables into Word and Copy for the results sentence.
How to report choice-based conjoint in APA style
Report the coefficients, part-worths, attribute importance and willingness to pay with the model fit.
Frequently asked questions
How is willingness to pay calculated?
WTP for a level = −β_level / β_price: the price change that would exactly offset the utility gain of that level.
What is the difference between CBC and rating conjoint?
CBC asks for choices, which mimic real purchase decisions; rating conjoint asks for scores of single profiles.
What does McFadden’s R² mean here?
1 − LL_model / LL_null, where the null model gives every alternative an equal chance; values of .2 to .4 indicate a good fit.
Related tools
Browse all categories
Data cleaning and preparation · Describe and explore · Compare means · Nonparametric tests · Correlation and regression · Bayesian statistics · Categorical data · Reliability and agreement · Multivariate analysis · Structural equation modeling · Econometrics and time series · Panel data and causal inference · Marketing and decision research · Survival, clinical and meta-analysis · Study planning and calculators · PLS-SEM