# How to Write a Reproducible Methods Section for Biological Experiments ## Quick Answer - A reproducible methods section requires explicit detail on experimental units, randomization, blinding, sample size, and replication so another laboratory can repeat the work without contacting the authors. - Use reporting guidelines from the [EQUATOR Network](https://www.equator-network.org/) to structure your methods and ensure all essential elements are documented before submission. - No template can guarantee reproducibility if you omit negative results, failed attempts, or equipment-specific quirks that affect how the protocol performs in practice. ## Why Reproducibility Fails in Biological Methods Sections The methods section is the most frequently skipped portion of a published paper, yet it determines whether another laboratory can verify your findings. When a researcher cannot reproduce your results, the failure often traces back to missing procedural details instead of flawed science. A 2016 survey of 1,576 researchers found that more than 70% had tried and failed to reproduce another scientist's experiment, and over half could not reproduce their own experiments. The most cited reasons were insufficient detail in methods descriptions and uncontrolled variables. Biological systems present unique challenges for reproducibility. Living organisms respond to environmental conditions, genetic variation, and subtle differences in handling that do not affect chemical or physical measurements. A cell line that grows normally in one laboratory may behave differently in another because of media composition, incubator temperature gradients, or passage number. An animal model may show different responses based on housing density, light cycles, or microbiome status. These variables are often invisible in the final publication because the methods section did not record them. The consequences of irreproducible methods extend beyond wasted time and resources. When other laboratories cannot repeat your work, the scientific record becomes unreliable. Clinical applications built on unreproducible findings can fail in translation, and public trust in research erodes. The [Committee on Publication Ethics](https://publicationethics.org/core-practices) core practices emphasize that authors must provide sufficient detail for others to evaluate and reproduce their work, and journals increasingly require methods sections that meet this standard. ## Core Principles of a Reproducible Methods Section ### Experimental Units and Their Boundaries The experimental unit is the smallest entity that receives an independent treatment assignment. In cell culture, this is typically a well, dish, or flask. In animal studies, it is the individual animal. In field experiments, it may be a plot or a cage. The methods section must state the experimental unit explicitly because it determines how data are analyzed and how many replicates are needed. A common error is treating subsamples as independent replicates. If you measure three cells from the same well, those measurements are not independent because they share the same treatment environment. The well is the experimental unit, and the three cells are technical replicates. Confusing these leads to inflated sample sizes and false confidence in the results. The methods section should state the experimental unit and explain how technical replicates were distinguished from biological replicates. ### Randomization and Its Role in Bias Prevention Randomization assigns experimental units to treatment groups so that uncontrolled variables are distributed evenly across groups. Without randomization, systematic differences between groups can be mistaken for treatment effects. For example, if all control animals are housed on the top shelf and all treated animals on the bottom shelf, temperature or light differences could confound the results. The methods section should describe the randomization procedure in enough detail that another researcher could repeat it. This includes the method used to generate the random sequence, such as a random number generator or a random number table, and the point at which randomization occurred. If randomization was stratified by a variable such as sex or weight, this should be stated. The [National Library of Medicine](https://www.ncbi.nlm.nih.gov/books) provides access to research methods references that describe randomization procedures in detail. ### Blinding and the Prevention of Observer Bias Blinding prevents the researcher from knowing which treatment a subject received, which reduces the risk of bias in data collection and analysis. In biological experiments, blinding is often difficult because treatments may have visible effects. However, the methods section should describe any blinding procedures that were used, including who was blinded and at what stage of the experiment. The absence of blinding is not necessarily a fatal flaw, but it should be acknowledged. The methods section should state whether blinding was used and, if not, why it was not feasible. This transparency allows readers to assess the risk of bias in the results. ### Sample Size Determination and Statistical Power The sample size is the number of experimental units per treatment group. It must be large enough to detect a meaningful effect if one exists. The methods section should describe how the sample size was determined, including the expected effect size, the desired statistical power, and the significance level. If a pilot study was used to estimate the effect size, this should be stated. A common failure is choosing a sample size based on convenience or precedent instead of statistical calculation. The methods section should justify the sample size with a power analysis or a reference to a published study with a similar effect size. If the sample size was limited by practical constraints, this should be stated as a limitation. ### Replication and the Distinction Between Technical and Biological Replicates Replication is the number of independent experimental units per treatment group. Biological replicates are independent samples from different organisms or cell culture preparations. Technical replicates are repeated measurements of the same sample. The methods section should state the number of biological and technical replicates and how they were distinguished. The distinction matters because biological replicates capture the natural variation in the system, while technical replicates capture measurement error. A study with many technical replicates but few biological replicates may have precise measurements but limited generalizability. The methods section should report both types of replication and the statistical analysis that accounts for the difference. ## The Reproducible Methods Template The following template provides a structured approach to writing a methods section that includes all the details necessary for another laboratory to replicate the experiment. Each section includes a fill-in-the-blank format and examples from published papers. ### Experimental Design Overview The experimental design section should provide a high-level summary of the experiment, including the hypothesis, the treatment groups, and the primary outcome. This section should be written so that a reader can understand the overall structure of the experiment before reading the detailed protocols. **Template:** - Hypothesis: [State the specific hypothesis being tested] - Treatment groups: [List each treatment group and the control group] - Primary outcome: [State the primary outcome measure] - Experimental unit: [State the experimental unit, such as a well, animal, or plot] - Number of replicates: [State the number of biological replicates per group] - Duration of experiment: [State the duration of the experiment] **Example:** - Hypothesis: Treatment with compound X reduces cell proliferation in a dose-dependent manner. - Treatment groups: Vehicle control, 1 µM X, 10 µM X, 100 µM X. - Primary outcome: Cell proliferation measured by EdU incorporation. - Experimental unit: Individual well in a 96-well plate. - Number of replicates: 6 biological replicates per group. - Duration of experiment: 72 hours. ### Experimental Units and Sample Size This section describes the experimental units in detail and justifies the sample size. It should include the source of the experimental units, the criteria for inclusion and exclusion, and the method for determining the sample size. **Template:** - Source of experimental units: [Describe the source, such as a cell line from a repository or animals from a breeding colony] - Inclusion criteria: [State the criteria for including an experimental unit in the study] - Exclusion criteria: [State the criteria for excluding an experimental unit from the study] - Sample size determination: [Describe the method used to determine the sample size, including the expected effect size, power, and alpha] - Sample size: [State the final sample size per group] **Example:** - Source of experimental units: MCF7 cells obtained from the American Type Culture Collection (ATCC). - Inclusion criteria: Cells were used at passage 5 to 10. - Exclusion criteria: Cells were excluded if the viability was below 90% at the start of the experiment. - Sample size determination: Sample size was calculated using a two-sided t-test with an alpha of 0.05, a power of 0.80, and an expected effect size of 0.5 standard deviations. - Sample size: 6 biological replicates per group. ### Randomization and Allocation This section describes how experimental units were assigned to treatment groups. It should include the method of randomization, the person who performed the randomization, and the point at which randomization occurred. **Template:** - Randomization method: [Describe the method, such as a random number generator or a random number table] - Randomization sequence: [Describe how the sequence was generated and stored] - Allocation concealment: [Describe how the allocation was concealed from the researcher] - Implementation: [Describe who performed the randomization and when] **Example:** - Randomization method: A random number generator was used to assign wells to treatment groups. - Randomization sequence: The sequence was generated using a computer program and stored in a password-protected file. - Allocation concealment: The allocation was concealed from the researcher who performed the treatment. - Implementation: The randomization was performed by a researcher who was not involved in the experiment. ### Blinding Blinding prevents bias in data collection and analysis. The methods section must describe who was blinded and at what stage of the experiment. **Template:** - Blinding of the researcher: [State whether the researcher was blinded to the treatment allocation] - Blinding of the data analyst: [State whether the data analyst was blinded to the treatment allocation] - Blinding of the outcome assessor: [State whether the outcome assessor was blinded to the treatment allocation] - Method of blinding: [Describe how blinding was achieved, such as using coded labels] - Unblinding: [Describe when and how the blinding was broken] **Example:** - Blinding of the researcher: The researcher who performed the cell culture was blinded to the treatment allocation. - Blinding of the data analyst: The data analyst was blinded to the treatment allocation. - Blinding of the outcome assessor: The outcome assessor was blinded to the treatment allocation. - Method of blinding: Treatment groups were labeled with codes that were not revealed until after the data analysis. - Unblinding: The codes were revealed after the data analysis was complete. ### Sample Size and Statistical Power This section describes the statistical considerations that determined the sample size. It must include the expected effect size, the desired power, and the alpha level. **Template:** - Expected effect size: [State the expected effect size, such as a difference in means or a fold change] - Desired power: [State the desired statistical power, such as 0.80] - Alpha level: [State the alpha level, such as 0.05] - Sample size calculation: [Describe the method used to calculate the sample size, such as a t-test or an ANOVA] - Sample size per group: [State the final sample size per group] **Example:** - Expected effect size: A 20% reduction in cell proliferation. - Desired power: 0.80. - Alpha level: 0.05. - Sample size calculation: Sample size was calculated using a two-sided t-test with an alpha of 0.05 and a power of 0.80. - Sample size per group: 6 biological replicates per group. ### Replication and Technical Replicates This section describes the number of biological and technical replicates and how they were distinguished. **Template:** - Biological replicates: [State the number of biological replicates per group] - Technical replicates: [State the number of technical replicates per sample] - Distinction: [Describe how biological and technical replicates were distinguished] **Example:** - Biological replicates: 6 biological replicates per group. - Technical replicates: 3 technical replicates per biological replicate. - Distinction: Biological replicates were independent wells in the same plate, and technical replicates were repeated measurements from the same well. ### Materials and Reagents This section lists all materials and reagents used in the experiment, including the source, catalog number, and lot number. This information allows another researcher to obtain the same materials and avoid variations between lots. **Template:** - Reagent: [Name of the reagent] - Source: [Name of the supplier] - Catalog number: [Catalog number] - Lot number: [Lot number] - Concentration: [Concentration used] **Example:** - Reagent: Dulbecco's Modified Eagle Medium (DMEM) - Source: Thermo Fisher Scientific - Catalog number: 11965-092 - Lot number: 1234567 - Concentration: 500 mL ### Equipment and Software This section describes the equipment and software used in the experiment, including the model, manufacturer, and settings. This information is essential for reproducing the experiment because equipment-specific settings can affect the results. **Template:** - Equipment: [Name of the equipment] - Model: [Model number] - Manufacturer: [Manufacturer] - Settings: [Settings used] **Example:** - Equipment: Cell culture incubator - Model: Heracell VIOS 160i - Manufacturer: Thermo Fisher Scientific - Settings: 37°C, 5% CO2 ### Statistical Analysis This section describes the statistical methods used to analyze the data, including the software, the statistical tests, and the criteria for significance. **Template:** - Software: [Name of the statistical software] - Version: [Version number] - Statistical tests: [List the statistical tests used] - Criteria for significance: [State the alpha level and the criteria for significance] - Post-hoc tests: [Describe any post-hoc tests used] **Example:** - Software: GraphPad Prism - Version: 9.0 - Statistical tests: One-way ANOVA with Tukey's post-hoc test. - Criteria for significance: P < 0.05. - Post-hoc analysis: Tukey's HSD test was used to compare all pairs of groups. ## At a Glance: Key Elements of a Reproducible Methods Section | Element | What to Include | Why It Matters | | --- | --- | --- | | Experimental unit | The entity that receives the treatment independently | Determines the statistical analysis and the number of replicates | | Randomization | The procedure for assigning experimental units to groups | Prevents systematic bias between groups | | Blinding | Who was blinded and how | Prevents bias in data collection and analysis | | Sample size | The number of experimental units per group and the justification | Ensures the study has sufficient power to detect an effect | | Replication | The number of biological and technical replicates | Distinguishes biological variation from measurement error | | Materials and reagents | Source, catalog number, and lot number | Allows other researchers to obtain the exact materials | | Equipment and settings | Model, manufacturer, and settings | Ensures the experiment can be repeated with the same equipment | | Statistical analysis | Software, tests, and criteria for significance | Allows other researchers to analyze the data in the same way | ## Practical Implementation Steps ### Step 1: Select the Appropriate Reporting Guideline The [EQUATOR Network](https://www.equator-network.org/) provides a searchable database of reporting guidelines for different study types. For biological experiments, the relevant guidelines include ARRIVE for animal research, CONSORT for clinical trials, and the guidelines for cell biology experiments. Select the guideline that matches your study type and use it as a checklist for your methods section. ### Step 2: Write the Methods Section Before the Experiment The methods section should be written before the experiment begins, not after. This ensures that all necessary details are recorded and that the experiment is designed with reproducibility in mind. Use the template above to draft the methods section, and update it as the experiment progresses. ### Step 3: Record All Materials and Reagents Keep a detailed record of all materials and reagents, including the source, catalog number, and lot number. This information should be recorded in the methods section and in the laboratory notebook. If a reagent is discontinued or changed, the methods section should note the change and the potential impact on the results. ### Step 4: Document Equipment Settings and Calibration Record the model, manufacturer, and settings for all equipment used in the experiment. This includes the incubator temperature, the centrifuge speed, and the microscope settings. If the equipment is calibrated, record the calibration date and the calibration method. ### Step 5: Describe the Statistical Analysis in Detail The statistical analysis section should describe the software, the version, the statistical tests, and the criteria for significance. It should also describe any post-hoc tests and the method used to handle missing data. This information allows other researchers to verify the analysis and to apply the same methods to their own data. ### Step 6: Review the Methods Section Against the Reporting Guideline Before submission, review the methods section against the reporting guideline you selected in Step 1. Check that every element on the checklist is included and that the methods section is complete. This review can be done by a colleague who is not involved in the experiment. ## Records and Measurements ### Laboratory Notebooks The laboratory notebook is the primary record of the experiment. It should contain the methods, the data, and any observations that are not included in the methods section. The notebook should be written in ink, with no blank spaces, and should be signed and dated on each page. The notebook should be stored in a secure location and should be available for review by other researchers. ### Electronic Data Management Electronic data should be stored in a structured format that is accessible to other researchers. The [National Institutes of Health](https://sharing.nih.gov/data-management-and-sharing-policy) data management and sharing policy requires that data be shared in a way that is findable, accessible, interoperable, and reusable. This includes storing the data in a public repository, providing metadata, and using a data management plan. ### Data Management Plans A data management plan describes how the data will be collected, stored, and shared. The [NIH](https://sharing.nih.gov/data-management-and-sharing-policy) requires a data management plan for all research that it funds. The plan should describe the types of data, the data formats, the metadata, the storage and backup procedures, and the data-sharing plan. ### Metadata and Documentation Metadata is the information that describes the data, such as the experimental conditions, the equipment settings, and the data collection methods. Metadata should be recorded in a structured format that is machine-readable. This allows other researchers to understand the data and to reproduce the experiment. ## Common Failure Patterns ### Missing Negative Results Negative results are often omitted from the methods section because they are not considered important. However, negative results are essential for reproducibility because they show what did not work. If a particular protocol failed, the methods section should describe the failure and the reason for the failure. This information helps other researchers avoid the same mistake. ### Incomplete Reagent Information Reagent information is often incomplete in the methods section. The source, catalog number, and lot number are essential for reproducibility because the same reagent from a different lot may behave differently. The methods section should include this information for every reagent. ### Vague Statistical Descriptions The statistical analysis is often described in vague terms, such as "statistical analysis was performed using a t-test." This does not provide enough information for another researcher to reproduce the analysis. The methods section should describe the statistical software, the version, the statistical tests, the criteria for significance, and the post-hoc tests. ### Inadequate Randomization and Blinding Randomization and blinding are often not described in the methods section. This makes it impossible for the reader to assess the risk of bias in the study. The methods section should describe the randomization procedure and the blinding method, or state that they were not used. ### Overlooking the Experimental Unit The experimental unit is often not defined in the methods section. This makes it difficult to determine the sample size and the statistical analysis. The methods section should state the experimental unit explicitly. ## Limitations and Interpretation ### The Limits of the Template The template provides a structured approach to writing a methods section, but it cannot guarantee reproducibility. The template does not cover every possible experimental design, and it does not capture the tacit knowledge that is often required to perform an experiment. The template should be used as a starting point, and the methods section should be adapted to the specific experiment. ### The Role of the Researcher The researcher is the most important factor in reproducibility. The methods section can describe the experiment in detail, but it cannot capture the skill and experience of the researcher. The researcher should be aware of the limitations of the methods section and should be willing to share additional details with other researchers. ### The Need for a Culture of Reproducibility Reproducibility is beyond a matter of writing a good methods section. It is a culture that values transparency and rigor. The [Committee on Publication Ethics](https://publicationethics.org/core-practices) emphasizes the importance of data integrity and the responsibility of authors to provide sufficient detail for others to reproduce their work. The research community should promote a culture of reproducibility by rewarding transparency and by providing the resources for researchers to share their methods and data. ## Safety and Regulatory Context ### Biosafety and the Methods Section The methods section should describe the biosafety measures that were used in the experiment. This includes the biosafety level of the laboratory, the use of personal protective equipment, and the procedures for handling hazardous materials. The methods section should also describe the disposal of hazardous waste. ### Institutional Review and Approval The methods section should state whether the experiment was approved by the institutional review board or the animal care and use committee. This is required for research involving human subjects or animals. The methods section should include the approval number and the date of approval. ### Data Management and Sharing Policies The [NIH](https://sharing.nih.gov/data-management-and-sharing-policy) data management and sharing policy requires that data be shared in a way that is findable, accessible, interoperable, and reusable. The methods section should describe the data management plan and the data-sharing plan. This includes the data repository, the data format, and the metadata. ## Professional Escalation Criteria ### When to Seek Help If you are unable to reproduce a published method, you should seek help from the authors of the paper or from a colleague with experience in the field. The authors may be able to provide additional details that were not included in the methods section. If the method is still not reproducible, you should consider whether the method is flawed or whether the experiment was not described in sufficient detail. ### When to Report a Problem If you suspect that a published method is not reproducible, you should report the problem to the journal that published the paper. The journal may be able to publish a correction or a letter to the editor. You should also report the problem to the [Committee on Publication Ethics](https://publicationethics.org/core-practices) if you suspect that the paper is fraudulent. ### When to Seek Professional Advice If you are unsure about the reproducibility of a method, you should seek advice from a statistician or a methodologist. They can help you to assess the experimental design and the statistical analysis. They can also help you to determine whether the method is reproducible. ## A Decision Framework for Choosing the Right Level of Methodological Detail The template in the previous section tells you what to include, but it does not tell you how much detail is enough for each element. Overdescribing trivial steps wastes space and obscures the critical variables. Underdescribing a sensitive step makes replication impossible. This section provides a practical decision framework that helps you calibrate the level of detail for each component of your methods section based on how much that component can affect the outcome. ### The Sensitivity Principle The core idea is simple: the amount of detail you provide for any procedural step should be proportional to how much that step can change the result when performed differently. A step that produces a nearly identical outcome across a wide range of variations needs only a brief description. A step where small changes in timing, temperature, or technique produce large changes in the result needs exhaustive detail. To apply this principle, ask three questions for each step in your protocol: 1. What is the acceptable range of variation for this step? 2. What happens to the outcome when the step is performed at the edges of that range? 3. Can another researcher determine from your description whether they are inside or outside the acceptable range? If the answer to the third question is no, you need to add more detail. If the step tolerates wide variation and the outcome is insensitive to it, a short description is sufficient. ### A Worked Example of the Sensitivity Analysis Consider a simple cell viability assay. The steps include thawing cells, culturing them, treating them with a compound, and measuring viability. Each step has a different sensitivity profile. **Thawing cells.** The thawing procedure can vary in water bath temperature, duration, and the speed of transferring cells to warm media. Most cell lines tolerate a range of thawing conditions without affecting viability. A brief description such as "cells were thawed in a 37°C water bath and transferred to prewarmed media" is sufficient. **Plating density.** The number of cells plated per well directly affects the growth rate, the timing of confluence, and the response to treatment. A 20% difference in plating density can change the outcome of a proliferation assay. This step requires precise detail, including the cell count method, the counting instrument, the dilution scheme, and the final density per well. **Treatment duration.** The duration of compound exposure is a major determinant of the response. A 30-minute difference in a 24-hour treatment may not matter, but a 30-minute difference in a 2-hour treatment can change the result. The methods must state the exact treatment window and the tolerance for variation. **Viability measurement.** The choice of assay, the incubation time with the reagent, and the plate reader settings all affect the measured value. The methods must specify the reagent, the incubation time, the wavelength, and the plate reader model. The sensitivity analysis forces you to think about which steps are critical and which are not. It prevents you from writing a methods section that is either too sparse or unnecessarily long. ### A Decision Matrix for Detail Allocation The following matrix helps you decide how much detail to provide for each step. The rows are the sensitivity of the outcome to the step, and the columns are the complexity of the step. | Sensitivity of outcome to step | Low complexity step | High complexity step | | --- | --- | --- | | Low sensitivity | Brief description | Brief description with key settings | | Medium sensitivity | Standard detail | Full detail with tolerances | | High sensitivity | Full detail with tolerances | Exhaustive detail with troubleshooting notes | For a low sensitivity and low complexity step, a single sentence is enough. For a high sensitivity and high complexity step, you should provide a step-by-step protocol with exact values, acceptable ranges, and notes on what can go wrong. ### The Tolerance Statement A tolerance statement is a sentence that tells the reader how much variation is acceptable for a given step. This is a powerful tool for reproducibility because it tells the reader also what you did but also how much they can deviate before the result is compromised. For example, instead of writing "cells were incubated for 24 hours," you can write "cells were incubated for 24 hours, with a tolerance of plus or minus 30 minutes." This tells the reader that a 23.5 hour incubation is acceptable but a 20 hour incubation is not. Tolerance statements are especially useful for steps that are time sensitive, temperature sensitive, or concentration sensitive. They also help you avoid the common failure of over specifying a step that does not need it. ### The Decision Log A decision log is a record of the choices you made during the experiment and the reasons for those choices. It is not part of the published methods section, but it is the source material for the methods section. The log should contain the following for each step: - The step name and the protocol used - The exact values for all parameters - The tolerance for each parameter - The reason for choosing the value - Any observations about how the step performed The decision log serves two purposes. First, it forces you to think through the sensitivity of each step before you write the methods. Second, it provides a record that you can consult when you are writing the methods section or when another researcher asks for additional details. ### The Verification Protocol A verification protocol is a set of checks you perform to confirm that your methods section contains enough detail for another researcher to replicate the experiment. The protocol has three stages. **Stage 1: The self-test.** After you write the methods section, read it as if you were a researcher who has never performed the experiment. For each step, ask whether you could perform the step with the information provided. If you cannot, add the missing detail. **Stage 2: The colleague test.** Ask a colleague who is not involved in the experiment to read the methods section and to identify any steps that are unclear or incomplete. The colleague should not be an expert in the specific technique, because the goal is to test whether the methods are clear to a general researcher. **Stage 3: The dry run.** If the experiment is simple enough, perform a dry run of the methods section in a different laboratory or with a different researcher. This is the most rigorous test of reproducibility, but it is not always feasible. If a dry run is not possible, the colleague test is the best alternative. ### The Detail Budget A detail budget is a practical tool for managing the length of your methods section. It allocates a certain number of words to each step based on its sensitivity. The budget prevents you from spending too many words on a low sensitivity step and too few on a high sensitivity step. To create a detail budget, list all the steps in the experiment and assign each step a sensitivity rating. Then allocate the total word count of the methods section proportionally to the sensitivity ratings. For example, if you have 1,000 words for the methods section and you have five steps with sensitivity ratings of 1, 2, 3, 4, and 5, you would allocate 67, 133, 200, 267, and 333 words to each step. The detail budget is not a rigid rule, but it is a useful check. If you find that you have written 300 words for a low sensitivity step and 50 words for a high sensitivity step, you should rebalance the allocation. ### The Reproducibility Score A reproducibility score is a simple checklist that you can use to assess the completeness of your methods section. It is not a substitute for the reporting guidelines from the [EQUATOR Network](https://www.equator-network.org/), but it is a quick check that you can perform before you submit your paper. The score is the percentage of the following items that you have included in your methods section: - Experimental unit stated - Sample size stated with justification - Randomization method described - Blinding method described or the absence of blinding stated - Biological and technical replicates distinguished - Source, catalog number, and lot number for all reagents - Equipment model, manufacturer, and settings - Statistical software, version, and tests - Tolerance statements for high sensitivity steps - Verification protocol performed A score of 100% means that you have included all the items. A score below 80% means that you should revise the methods section before submission. The score is a practical tool for self-assessment, and it is not a substitute for the formal reporting guidelines. ### The Revision Loop The revision loop is the process of updating the methods section as the experiment progresses. You should not write the methods section once and then forget about it. Instead, you should update it whenever you change a parameter, discover a new variable, or encounter a problem. The revision loop has four steps: 1. Record the change in the decision log 2. Assess the sensitivity of the change 3. Update the methods section with the new detail 4. Recalculate the reproducibility score The revision loop ensures that the methods section reflects the experiment as it was actually performed. It also ensures that you do not forget to include a change that you made during the experiment. ### The Comparison Table The following table compares the traditional approach to writing a methods section with the decision framework described in this section. | Aspect | Traditional approach | Decision framework approach | | --- | --- | --- | | Detail level | Uniform for all steps | Proportional to sensitivity | | Tolerance | Not stated | Stated for key variables | | Verification | None | Three stage protocol | | Length | Uncontrolled | Managed with a detail budget | | Update | Written once | Updated through a revision loop | | Self assessment | None | Reproducibility score | The decision framework approach is more work than the traditional approach, but it produces a methods section that is more likely to be reproducible. The framework also helps you to write the methods section more efficiently because you do not waste words on steps that do not need them. ### The Role of the Decision Framework in the Publication Process The decision framework is not a replacement for the reporting guidelines from the [EQUATOR Network](https://www.equator-network.org/). The reporting guidelines tell you what to include, and the decision framework tells you how much detail to include for each element. You should use both together. The decision framework also supports the data management and sharing expectations of the [National Institutes of Health](https://sharing.nih.gov/data-management-and-sharing-policy). The decision log and the verification protocol are part of the documentation that you should keep with your data. This documentation helps other researchers to understand the choices you made and to assess the reproducibility of your work. The [Committee on Publication Ethics](https://publicationethics.org/core-practices) core practices emphasize the importance of accurate and complete reporting. The decision framework is a practical way to meet this expectation. It helps you to write a methods section that is complete, accurate, and reproducible. ### The Limits of the Decision Framework The decision framework is a practical tool, but it has limits. It cannot tell you the exact sensitivity of a step without experimental data. You may need to perform pilot experiments to determine the tolerance for a step. The framework also cannot capture the tacit knowledge that is often required to perform a biological experiment. Some steps are difficult to describe in words, and the framework does not solve this problem. The framework also assumes that you have the time to perform the sensitivity analysis and the verification protocol. In practice, you may not have the time to perform a full verification for every experiment. In this case, you should prioritize the steps that are most likely to affect the outcome and focus your verification effort on those steps. ### The Practical Implementation of the Decision Framework To implement the decision framework in your own work, follow these steps: 1. List all the steps in your experiment 2. Rate the sensitivity of each step on a scale of 1 to 5 3. Create a detail budget based on the sensitivity ratings 4. Write the methods section using the detail budget 5. Add tolerance statements for the high sensitivity steps 6. Perform the verification protocol 7. Calculate the reproducibility score 8. Update the methods section through the revision loop The decision framework is a skill that improves with practice. The first time you use it, it may take longer than your usual approach. After you have used it for a few experiments, it becomes a habit, and the time cost decreases. ### The Relationship to the Reproducible Methods Template The decision framework is a complement to the reproducible methods template in the previous article. The template provides the structure for the methods section, and the decision framework provides the method for deciding how much detail to include in each part of the template. You should use the template to organize the methods section and the decision framework to calibrate the detail. The template and the decision framework together form a complete system for writing a reproducible methods section. The template ensures that you do not forget any essential element, and the decision framework ensures that you do not waste words on unnecessary detail. The result is a methods section that is complete, accurate, and reproducible. ## Frequently Asked Questions ### What is the difference between a biological replicate and a technical replicate? A biological replicate is an independent measurement of a biological sample, such as a cell from a different well or a different animal. A technical replicate is a repeated measurement of the same sample. Biological replicates capture the natural variation in the system, while technical replicates capture the measurement error. The methods section should state the number of biological and technical replicates and how they were distinguished. ### How do I determine the sample size for my experiment? The sample size is determined by the expected effect size, the desired statistical power, and the alpha level. The expected effect size is the difference between the treatment groups that you want to detect. The desired power is the probability of detecting the effect if it exists. The alpha level is the probability of a false positive. The sample size can be calculated using a statistical software or a sample size calculator. ### What is the purpose of randomization in an experiment? Randomization assigns experimental units to treatment groups in a way that is not influenced by the researcher. This prevents the bias from affecting the results. Randomization ensures that the treatment groups are similar at the start of the experiment, so that any differences at the end of the experiment can be attributed to the treatment. ### What is blinding and why is it important? Blinding is the practice of keeping the researcher unaware of the treatment allocation. This prevents the bias from affecting the data collection and analysis. Blinding is important because the researcher may unconsciously influence the results if they know which treatment a subject received. ### What is the role of the reporting guideline in the methods section? A reporting guideline is a checklist of the information that should be included in a publication. The [EQUATOR Network](https://www.equator-network.org/) provides a database of reporting guidelines for different study types. The reporting guideline helps the researcher to ensure that the methods section includes all the necessary information for reproducibility. ### How do I describe the equipment and software in the methods section? The methods section should describe the equipment and software in detail, including the model, the manufacturer, and the settings. This information allows other researchers to obtain the same equipment and to use the same settings. The methods section should also describe the calibration of the equipment. ### What should I do if I cannot reproduce a published method? If you cannot reproduce a published method, you should first ask the authors for additional details. If the method is still not reproducible, you should consider whether the method is flawed or if the experiment was not described in enough detail. You should also report the problem to the journal that published the paper. ### How do I write a data management plan? A data management plan describes how the data will be collected, stored, and shared. The plan should include the types of data, the data formats, the metadata, the storage and backup, and the sharing plan. The [NIH](https://sharing.nih.gov/data-management-and-sharing-policy) requires a data management plan for all research that it funds. ## Related Diagnostic Guides - [Sequencing Primer Design: Key Considerations for Reliable Results](/knowledge/diagnostics/molecular/sequencing-primer-design-key-considerations-for-reliable-results) - [Animal Biology](/blog/guides/animal-biology) - [Animal Research](/blog/guides/animal-research) ## References and Further Reading - [Research Methods Resources](https://www.ncbi.nlm.nih.gov/books). National Library of Medicine. - [EQUATOR Network](https://www.equator-network.org/). EQUATOR Network. - [Core Practices](https://publicationethics.org/core-practices). Committee on Publication Ethics. - [NIH Grants and Funding](https://grants.nih.gov/). National Institutes of Health. - [ORCID for Researchers](https://info.orcid.org/researchers). ORCID. - [Data Management and Sharing Policy](https://sharing.nih.gov/data-management-and-sharing-policy). National Institutes of Health. - [Laboratory Biosafety Manual](https://www.who.int/publications/i/item/9789240011311). World Health Organization. - [NCBI Literature Resources](https://www.ncbi.nlm.nih.gov/). National Center for Biotechnology Information. - [Cell-Free Protein Synthesis-Based Biosensing Platforms for Clinical Diagnostics.](https://pubmed.ncbi.nlm.nih.gov/41979265). ACS sensors, 2026. - [Proteoform-Specific Insights into Cellular Proteome Regulation.](https://pubmed.ncbi.nlm.nih.gov/27451424). Molecular & cellular proteomics : MCP, 2016. - [Blind spots on western blots: Assessment of common problems in western blot figures and methods reporting with recommendations to improve them.](https://pubmed.ncbi.nlm.nih.gov/36095010). PLoS biology, 2022. > This article is educational and does not replace validated laboratory procedures, institutional biosafety review, manufacturer instructions, or professional interpretation.