Social Sciences and Humanities metadata

Unit of Analysis 

Optional

Basic unit of analysis or observation that describes the dataset.

Repeatable.

They can be individuals, families, groups, institutions/organizations, administrative units and others. The vocabulary controlled by this field can be consulted on the DDI Alliance website, under the Analysis Unit section.

• Individual 
• Organization 
• Family 

Universe

Optional

Description of the population covered by the data: the group of people or other elements that are the subject of the study and to which the results of the study refer.

Repeatable.

Age, nationality, and residency usually help delineate a particular Universe, but other factors may be used, such as age limits, sex, marital status, race, ethnic group, nationality, income, criminal convictions, and more. In general, from the description of the Universe it should be possible to determine whether an individual or element is a member of the population under study.

• Spaniards between 12 and 30 years old.
• Second year medical students at the University of Barcelona.

Time Method

Optional 

The temporal method or temporal dimension of data collection.

Fill in information such as cross section, trend, time series or others.

• Longitudinal
• Temporal series

Data Collector

Person, agency or institution responsible for interviews or questionnaires in data collection.

In the personal name it is necessary to follow the form: Surname, Name.

Non-Roman names can be translated according to ALA-LC12 schemes.

For institutions, indicate the developed name of the institution. If, in addition, a structural unit must be added (department, research group, etc.) it must be indicated after the name followed by a period.

• Charpy, Antoine 
• Universitat de Barcelona. Department of Chemistry

Collector Training

Optional 

Type of training provided to the data collector.

Free text.

• The interviews were conducted by a professional interviewer under the supervision of the Social Science Research Institute.
• Each staff member received extensive training before starting work on the survey. Interviewers received about three days of classroom training in addition to self-training materials. Additional study materials and classroom training were planned during the interview period. Quality control measures such as editing results and observing interviews were used during the survey.

Frequency

Optional 

If the data collected includes more than one time point, indicate how often it was collected.

The frequency can be annual, quarterly, monthly, weekly, daily... Periodicity in data collection.

• Daily
  
• Weekly
  
• Monthly 

Sampling Procedure

Optional

Types of samples and sampling design used to select respondents to represent the population.

May include references to target sample size and sampling fraction.

• Probability
• Catalan adults selected at random.

Actual 

Optional 

Actual sample size.

The actual sample size of the research study may be entered in this field for reference purposes.

Enter a whole number, without decimals.

• 2000
• 50000 

Formula 

Optional 

Formula for determining target sample size.

A general plain text description of a sample size formula may be provided for reference purposes. It may include particular methodologies, practices and results from the existing scientific literature.

• Eligible employees who have lived in the following areas: K6V, K7A, K7C, K7G, K7H, K7P.

Major Deviations for Sample Design

Optional 

It shows the correspondences and discrepancies between the sample units (obtained) and the statistics available for the population (age, marital status...).

Free text.

• Oversampling in people over 50 years old.

Collection Mode

Optional

Method used to collect the data; instruments used.

Repeatable.

Characteristics of the instrumentation, such as telephone interview, online questionnaire or others.

• Telephone interviews
• Online questionnaires

Type of Research Instrument

Optional

Type of data collection instrument used.

Structured or "structured" indicates an instrument in which the same questions / tests are asked to all respondents, possibly with precoded answers. If a small part of this questionnaire includes open-ended questions, you must provide appropriate comments.

Semi-structured or "semi-structured" indicates that the research instrument mainly contains open questions.

Unstructured or “unstructured” indicates that in-depth interviews were conducted.

• Structured
• Semi-structured
• Unstructured

Characteristics of Data Collection Situation

Optional 

Description of the noteworthy aspects of the data collection situation.

It includes information on factors such as the cooperation of respondents, the length of interviews, the number of calls, or the like.

• There were 1419 respondents who answered questions in telephone interviews of approximately 35 minutes each. Clarifications to the survey questions were limited and respondents were asked to provide an answer based on the information provided so as not to allow interviewer biases/assumptions to influence the survey results.

Actions to Minimize Losses 

Optional 

Summary of actions taken to minimize data loss.

Include information about actions such as follow-up visits, monitoring controls, historical matching, estimation, etc.

• Reminder emails were sent to the target population.
• Cards reminding parents of the follow-up visit were given.

Control Operations 

Optional 

Methods to facilitate data control performed by the principal investigator or the data archive.

Free text.

• Field validation is built into data collection forms.
• Double-blind data entry and third-party cross-validation were used.

Weighting

Optional

The use of sampling procedures may make it necessary to apply weights to produce accurate statistical results.

Describe the criteria for using weights in the analysis of a collection. If a formula or weighting coefficient was developed, provide the formula, define its elements, and indicate how the formula was applied to the data.

• The final sample obtained for each area is not proportional to the population of Alberta. For example, Edmonton is oversampled, as shown in TABLE 1. Edmonton represents only 24% of Alberta's population, but has 43% of the interviews. Therefore, to combine the samples for a provincial sample, weighting is necessary. The weighting factors used for the 1987 survey are as follows: Edmonton 0.558439, Calgary 1.151521 and Other Alberta 1.471173.

Cleaning Operations

Optional 

Methods used to clean data.

Indicate if there has been a consistency check, code check or others.

• Physiological data were reviewed for outliers. Individual breaths with tidal volume (VT), respiratory rate (RR), or minute ventilation (ME) that were outside the 95% confidence interval for all infants were removed as outliers; 99.7% of all measured breaths were included in the final analyses.

Study Level Error Notes

Optional 

Any information that notes or clarifies the methodology and processing of the study.

Free text.

• The computerized quiz contains many features designed to
  maximize the quality of the data collected. Many edits are built into the questionnaire to check reported data for unusual values and detect logical inconsistencies. When an edit fails, the interviewer is asked to correct the information (with the help of the respondent, if necessary). Once the data has been transmitted to the central office, a series of processing steps are carried out in order to verify each questionnaire in detail. Invalid responses are corrected or flagged for imputation. The modifications were applied on a micro level. Deterministic and micro-level consistency edits were also performed. Data were checked for outliers and outliers and corrected at a micro level when necessary.

Response Rate 

Optional

Percentage of individuals in the sample who provided information.

Provide the contextualized percentage.

• At one month of follow-up: 60.2% (n=136/226).

Estimates of Sampling Error 

Optional

Measure of how accurately the population value can be estimated from a given sample.

Include confidence intervals, non-response and response bias.

• Plus or minus 2.5%, Weight design effect not calculated.
• In 2016, the 95% confidence interval for the median net worth of Canadian families could be quantified at $38,500.

Other Forms of Data Appraisal 

Optional 

Other matters relating to the evaluation of data.

Describe issues such as response variance, non-response rate and tests of bias, interviewer and response bias, confidence levels, question bias or the like.

• The OSBD is subject to interpretation as it is an indirect behavioral measure of perceived distress.

Type 

Optional 

Note type

Note types can be processing, system, observation, addendums, comments or others.

• Processing note

Subject 

Optional 

Subject of the note.

What is the note about?

• Corrections in the variables

Text 

Optional 

Note text.

This note must be related to the fields of the Social Sciences and Humanities section.

• Note information (Misc) v2: Corrections were made to the variables: PAS1MRG1, PAS1MRG2, PASRDPO1, PASRDPO2, PASRDPO3, PASRDPO4, PASRDPO5 and VERDATE.