Plain-language reference

Survey, research, analytics, data, and security glossary

Understand the terms used across SierraTec Survey and professional feedback programs. Definitions are educational summaries and should be interpreted in the context of your method, policy, contract, and jurisdiction.

Showing 160 terms

A

A/B test

A controlled comparison in which participants are randomly assigned to different versions to estimate the effect of a change.

A

Abandonment rate

The percentage of started survey sessions that do not reach the defined completion point.

A

Anonymous response

A response collected without identifiers that the survey owner or platform can reasonably use to link it to a person. The label should not be used when identity can still be inferred.

A

API

An application programming interface that allows software systems to exchange data and trigger operations through documented requests.

A

Audit log

A time-ordered record of security-relevant or administrative actions, including actor, target, result, context, and time.

B

Benchmark

A reference value used for comparison, such as a prior period, peer group, target, or external norm.

B

Bias

A systematic influence that moves results away from the value or interpretation the study intends to estimate.

B

Branching

Survey logic that sends respondents to different pages or endings based on answers or other conditions.

C

CAPTCHA

A challenge used to reduce automated submissions. It can support abuse prevention but does not guarantee a human or unique respondent.

C

Census

An attempt to collect information from every member of a defined population rather than a sample.

C

Closed-ended question

A question that asks respondents to choose from predefined answer options.

C

Coding

The process of assigning categories or numeric values to answers so they can be analyzed consistently.

C

Completion rate

The percentage of survey starts or eligible sessions that reach the defined completion point.

C

Confidence interval

A range calculated from sample data that reflects uncertainty around an estimate under stated assumptions.

C

Confidence level

The long-run percentage of similarly constructed confidence intervals expected to contain the population value under the model assumptions.

C

Constant-sum question

A question that asks respondents to allocate a fixed total, often 100 points, across options.

C

Construct

An abstract concept such as trust, satisfaction, engagement, or effort that is measured using one or more observable items.

C

Correlation

A measure of association between variables. Correlation alone does not establish that one variable causes another.

C

Cross-tabulation

A table that compares the distribution of one variable across categories of another variable.

C

CSAT

Customer Satisfaction Score, commonly calculated as the percentage of respondents selecting specified positive options on a satisfaction scale.

C

Customer Effort Score

A measure of how easy or difficult customers report that an interaction or task was, using a defined question and scale.

D

Data cleaning

A documented process for identifying and handling missing, duplicate, inconsistent, invalid, or otherwise problematic records.

D

Data minimization

Collecting, using, and retaining only the personal or sensitive information necessary for a defined purpose.

D

Data retention

The policy and operational period for keeping survey content, responses, files, logs, and backups before deletion or archive.

D

Detractor

In standard NPS grouping, a respondent who selects a score from 0 through 6.

D

Display logic

A rule that shows or hides a question, answer, or content block when specified conditions are met.

D

Distribution channel

The method used to reach respondents, such as email, public link, QR code, SMS, embed, kiosk, or authenticated invitation.

D

Dropout

A respondent who begins but does not complete the survey or study according to the defined completion rule.

D

Duplicate response

A submission believed to come from the same respondent, invitation, account, browser, or source as another submission under a defined rule.

E

Embedded survey

A survey displayed within another website or application rather than opened as a separate standalone page.

E

Encryption at rest

Cryptographic protection applied to stored data and media, together with controlled key management.

E

Encryption in transit

Cryptographic protection for data moving between systems, commonly provided by TLS.

F

Finite population correction

An adjustment that can reduce a required sample size when sampling without replacement from a known, relatively small population.

F

Forced-choice question

A question that requires respondents to select an offered option and may omit a neutral, not-applicable, or opt-out choice.

F

Frequency distribution

A count or percentage showing how often each value or category appears.

G

GDPR

The European Union General Data Protection Regulation. Applicability and obligations depend on the processing context and should be assessed by qualified professionals.

I

Identified response

A response intentionally connected to a known person, account, invitation, or other direct identifier.

I

Idempotency

A property that allows the same operation to be retried without creating unintended duplicate effects.

I

Invitation collector

A survey collection method that sends or assigns a unique invitation to a known contact or participant.

L

Likert item

A statement followed by ordered response choices, often from strongly disagree to strongly agree.

L

Likert scale

A set of related Likert items intended to measure a construct. The term is often used informally for a single agreement item.

L

Longitudinal study

A study that collects data from the same or comparable populations at multiple points over time.

M

Margin of error

A sample-based expression of statistical precision around an estimate at a stated confidence level and under specified assumptions.

M

Matrix question

A compact grid that asks respondents to rate multiple rows using the same set of columns or scale.

M

Mean

The arithmetic average calculated by summing numeric values and dividing by the number of included values.

M

Median

The middle value after observations are ordered, or the average of the two middle values when the count is even.

M

Metadata

Information about a response or event, such as timestamp, collector, language, device class, campaign, or operational identifiers.

M

Mixed methods

A research approach that intentionally combines quantitative and qualitative data and analysis.

M

Multi-factor authentication

Authentication that requires evidence from more than one factor category, such as password plus authenticator app.

N

Net Promoter Score

A score from -100 to 100 calculated as the percentage of promoters minus the percentage of detractors using the standard 0-10 recommendation item.

N

Nonresponse bias

Bias that occurs when respondents and nonrespondents differ in ways related to the survey measures.

O

Open-ended question

A question that lets respondents provide an answer in their own words rather than selecting only predefined options.

O

OpenID Connect

An identity layer built on OAuth 2.0 that allows applications to authenticate users and receive verified identity claims.

O

Outlier

An observation that is unusually distant from other values and may represent valid variation, error, or a special case requiring review.

P

Panel

A recruited group of people who have agreed to be invited to surveys or research studies over time.

P

Passive

In standard NPS grouping, a respondent who selects a score of 7 or 8.

P

Piping

Inserting a prior answer or known value into later survey text, answers, messages, or logic.

P

Population

The complete group about which a study intends to make statements or estimates.

P

Probability sampling

A sampling approach in which population members have known, nonzero selection probabilities.

P

Promoter

In standard NPS grouping, a respondent who selects a score of 9 or 10.

Q

Quota

A collection limit or target for a defined respondent group, answer, or characteristic.

R

Randomization

Using chance to vary question order, answer order, version assignment, or selection in order to reduce order effects or support comparison.

R

Ranking question

A question that asks respondents to order options by preference, importance, performance, or another criterion.

R

Rating scale

An ordered set of values used to indicate intensity, quality, agreement, likelihood, or another dimension.

R

Reliability

The consistency or stability of a measure under comparable conditions.

R

Representative sample

A sample whose composition and selection process support reasonable inference to the intended population for the measures of interest.

R

Response bias

Systematic distortion caused by how respondents interpret, remember, or choose to answer questions.

R

Response rate

The percentage of eligible invited or sampled units that provide a response under a clearly stated denominator and completion rule.

S

Sample

The subset of a population from which data are collected.

S

Sample size

The number of observations, respondents, or completed records included in a calculation or analysis.

S

Sampling error

Random difference between a sample estimate and the population value that arises because only a sample was observed.

S

SAML

Security Assertion Markup Language, commonly used for enterprise browser-based single sign-on between an identity provider and a service provider.

S

SCIM

System for Cross-domain Identity Management, a standard for automating user and group provisioning between identity systems and applications.

S

Sentiment analysis

A manual or automated classification of text by expressed tone or polarity. Results require validation and contextual review.

S

Skip logic

A rule that bypasses questions or pages based on prior answers or other conditions.

S

SPSS

Statistical software and associated data formats commonly used in social science and applied research workflows.

S

Standard deviation

A measure of how spread out numeric observations are around their mean.

S

Statistical significance

A model-based assessment of how compatible observed data are with a specified null hypothesis. It does not by itself establish practical importance or causation.

S

Stratified sampling

A probability sampling method that divides a population into strata and samples within each stratum.

S

Survey fatigue

Reduced willingness or care in responding because of survey frequency, length, repetition, or burden.

S

Survey weighting

Applying numeric weights so analyzed responses better reflect a target population or sampling design under explicit assumptions.

V

Validity

The degree to which evidence and theory support the intended interpretation and use of a measure or result.

W

Webhook

An HTTP notification sent by one system to another when a specified event occurs.

W

White label

Branding and domain controls that reduce or remove the platform provider's visible identity in the respondent or reporting experience.

Use terms carefully

Clear definitions improve survey decisions

Terms such as anonymous, representative, significant, confidential, compliant, and secure carry specific expectations. Use them only when the survey method and implementation support the claim.

Glossary Intelligence

Research and statistics terminology connected to real tools

The glossary now covers 160 terms across survey design, sampling, statistics, weighting, longitudinal research, qualitative work, privacy, security, identity, integrations, and governance.

75 new research/statistical terms

Power, effect size, regression, weighting, multiplicity, nonparametric tests, reliability, validity, longitudinal methods, reproducibility, and more.

Commonly confused

Compare standard deviation vs standard error, reliability vs validity, random sampling vs random assignment, and statistical vs practical significance.

Tool-linked definitions

Terms connect to SierraTec Stats, Sampling & Weighting Studio, longitudinal research, qualitative analysis, and survey guides where applicable.