upload assignment-2

02_activities/assignments/a2_survey_design_and_evaluation.md

@@ -21,7 +21,7 @@ Select one of the scenarios below and design a survey to meet the need(s) outlin
 
 For the **Canadian General Social Survey on Giving, Volunteering, and Participating, 2018 (cycle 33)**, conducted by Statistics Canada find any and all available documentation for the data gathered and identify and describe the survey features indicated below.
 
-1. Sample type
+1. Sample type: 
 2. Sample size
 3. Target population
 4. Sampling frame
@@ -36,34 +36,85 @@ For the **Canadian General Social Survey on Giving, Volunteering, and Participat
 13. Link to documentation and any additional sources used
 
 
+
 # Your Changes
 
 ## Part A - Survey Design: 
 
-The number of your chosen topic: `#`
+The number of your chosen topic: `Scenario 1`
 
 Describe the purpose of your survey:
 ```
 write your answer here...
 ```
+The purpose of the survey is to determine and understand the factors for the high turnover rate across multiple departments in the tech company. The survey focuses on sampling entry and lower-level positions and assessing what changes are required to improve employee satisfaction. 
 
 Describe your target population, sampling frame, sampling units, and observational units:
 ```
 write your answer here...
 ```
+The target population is all entry and lower-level positions across all departments in this tech company who are currently employed or who have recently left the company. These employees are the target population because they are directly affected by and contributing to the high turnover rate at the company. 
+The sampling frame is the company's Human Resources records, including employee databases that list all current entry and lower-level employees and employees that have had recent departures (within the last year). This frame provides an accessible list from which individuals can be selected for the study to determine employee satisfaction and reasons for turnover across departments. 
+
+The sampling units are individual employees listed in the Human Resources records/database who meet the criteria of being in entery or lower level roles in any department in the company. Each employee who meets the criteria represents one unit that can be selected into the sample. 
+
+The observational units are the individual employees whose data are being collected. Each observational unit can be provided through information collected through surveys, interviews, or exit questionnaires about job statisfaction, workplace conditions, management practices, compensation, and career development opportunities. 
+
+My overall sampling strategy would consist of using stratified random sampling. 
+The stratification variables would consist of department and tenure category (number of years worked within the department). 
+Sampling units would consist of individual employees that are currently working at the company and recently departed employees (within 6-12 months). 
+The sampling selection steps would consist of:
+- Dividing the population based on department and tenure. 
+- Use simple random sampling of the employees within each stratum. 
+- Include a separate sample of former employees from exit data from interviews/previous surveys using the same stratification. 
+Survey instrument: I would use an online anonymous questionnaire, so it is easily accessible. The survey would contain neutral phrasing and more close ended questions for sensitive questions. It will also include a N/A option, if respondents feel uncomfortable with answering certain questions. The anonymous factor will help maintain confidentiality and improve response rates. I would present the survey in meetings and have team leaders remind their departments to complete the survey to increase the response rate. 
+Survey timeframe: The survey would be open for 1 month. 
+Weighting scheme: I would use stratum-adjusted weights/calibrated weights for questions about job satisfaction, rentention/turnover intent, and turnover drivers (questions 3-10) to correct unequal selection probabilities across the stratas, adjust for non-response within the strats, and to match the HP population totals. The weighted scheme will help ensure consistency between the stratas, compare the results across all the questions, and avoid bias in oversampled stratas. 
 
 Your 5-10 question survey:
 ```
-1. write your question here...
-2. write your question here...
-3. write your question here...
-4. write your question here...
-5. write your question here...
-6. write your question here... (optional)
-7. write your question here... (optional)
-8. write your question here... (optional)
-9. write your question here... (optional)
-10. write your question here... (optional)
+1. What department do you currently work in or most recently worked in? 
+- Engineering 
+- Product 
+- IT 
+- Sales 
+- Marketing 
+- Customer Support
+- Operations 
+- Human Resources
+- Other (please specify) 
+
+2. How many years have you worked in the department? 
+- Less than 6 months
+- 6-12 months 
+- 1-2 years 
+- 2-3 years 
+- More than 3 years 
+
+Scale: 1 = Strongly Disagree, 2 = Disagree, 3 = Neutral, 4 = Agree, 5 = Strongly Agree, N/A (not applicable) 
+
+3. My day-to-day workload is manageable and expectations for my role are clearly communicated. 
+
+4. I receive adequate support, feedback, and guidance from my direct manager or supervisor. 
+
+5. I have opportunities for professional development, learning, or career advancement within the company. 
+
+6. My contributions are adequately recognized by my team or department. 
+
+7. My compensation and benefits are fair for my job position and market expectations. 
+
+8. Overall, how satisfied are you with your experience working at this company? 
+
+9. I am looking or have looked for different roles inside or outside of the company within the last 12 months. 
+
+10. Which of the following factors would impact your decision to leave the company? (select all that apply) 
+- Compensation
+- Workload/burnout
+- Management or leadership
+- Lack of growth or professional developmenet opportunities 
+- Work-life balance 
+- Company culture or team environment 
+- Other (please specify) 
 ```
 
 ## Part B - Survey Evaluation:
@@ -74,6 +125,66 @@ Identify and describe survey features:
 write your answer here
 ```
 
+1. Sample type: The survey uses a probability sample from a stratified cross-sectional design. The stratification was completed at the province/census metropolitan area (CMA) level. Rejective sampling was also used (volunteers vs. non-volunteers), sub-sampling was completed for respondents who were not volunteers. 
+
+2. Sample size: Approximately 50 000 units was used. Among the 50 000 units, about 40 000 were selected and mailed an invitation letter to complete the electronic questioinnaire. The expected completion number was 24 000 questionnaires. 
+
+3. Target population: The target population was all persons 15 years of age and older living in the ten provinces of Canada, excluding residents living in the 3 Territories and full-time residents of insitutions (residing for more than 6 months). 
+
+4. Sampling frame: Consists of a combination of landline and cellular telephone numbers from the Census and many administrative sources with Statistics Canada's dwlling frame.
+
+5. Survey mode(s): From the letter mailed out to the sample population, it asked participants to complete an electronic questionnaire (self-administered). Participants also had the option of completing the survey through CATI (computer assisted telephone interviewing) with the choice between French and English. 
+
+6. Timeline: Data was collected from September 4 to December 28, 2018.
+
+7. Response rate: The overall response rate was 41.9%. 
+
+8. Weights: WGHT_PER was used to analyze at the person level. Bootstrap weights were created for design-based variance estimation. Weights were also adjusted to match the population distributions and to correct for subsampling for non-volunteers. Finally, the survey weights were adjusted so that the weighted income distribution in the GVP data aligned with the 2017 CIS income distribution at the provincial level. The weights help ajust for the representation of the target population with certain characteristics. 
+
+9. Data processing: Processing followed the SSPE set of generalized steps and utilities. A structured environment was used to monitor the data processing. Family relationship consistency checks were conducted to ensure the integrity of the matrix data. Consistency and flow edits were conducted to ensure consistency of survey data and ensure participants used the correct path and fix off-path situations. The CATI system was used for error detection and to edit the flow of the questionnaire in real time. It would help identify out of range values and fix the issue with the participant. If the issue could not be fixed immediately, the interviewer would forward the data to head office for further review and editing. Manual and automated checks were conducted by health office after collection. 
+
+As part of data processing, Statistics Canada applied extensive validation and quality assurance procedures prior to data dissemination. These included analyses of changes over time, verification of estimates through cross-tabulations, and comparisons with other similar data sources to identify inconsistencies or anomalies. This part of the data processing can also apply to Cleaning but it mainly affects data processing.
+
+10. Cleaning, imputation, etc
+
+Item non-response was addressed via donor imputation, where missing values were replaced with values from similar “donor” records through a score function. Records with missing data were matched to similar complete records using a scoring system, and the closest matching or nearest donor was selected randomly if there was a tie to supply the missing values. When donor imputation was not feasible, mean imputation from a group of similar donors was used instead.
+
+Income data was linked to tax records for respondents who consented, reducing direct respondent burden and improving accuracy.
+
+Family income and personal income were imputed through using a direct linkage with a variable from the T1FF that corresponded with the census family income. 
+
+11. Sources of error: 
+Sampling error: inherent since the survey is based on a sample. Estimates will vary from sample to sample compared to results that would have been obtained from a complete census. 
+
+Coverage error: Occurs when there are differences between the target population and the surveyed population. Households without telephones or outside the frame may be under-represented.
+
+Non-response error: some units do not respond at household or individual levels, so the survey estimates were adjusted by using weighted responses. 
+
+Response and processing errors could also occur. 
+
+12. Limitations, known biases, etc
+
+Coverage would be considered a limitation as it may still omit certain segments of the target population (e.g., households without telephone numbers). The respondents could have changed their telephone number or have a different address, which could affect the number of respondents and response rate. As well CATI is only offered in English and French, so respondents with language barriers are less likely to complete the survey. The estimated average time to complete the survey was also estimated at 44 minutes which is quite long. It could lead to people not fully completing the survey. Estimates may differ from previous cycles due to the introduction of online collection, so respondents who do not have computers or have trouble using computers could affect the response rate. 
+
+Non-response bias from households with telephone services not covered by the current frame and the excluded target population remains possible despite weighting adjustments. 
+
+13. Link to documentation and any additional sources used
+
+Official survey documentation:
+
+General Social Survey (Cycle 33) — PUMF User Guide & Documentation:
+https://www150.statcan.gc.ca/n1/en/catalogue/45250011
+
+Data & files:
+
+Open data repository with questionnaire, codebooks, and user guide:
+https://hdl.handle.net/11272.1/AB2/GBFDYG
+
+Survey methodology & description:
+
+Statistics Canada survey overview (IMDB):
+https://www23.statcan.gc.ca/imdb/p2SV.pl?Function=getSurvey&Id=796234
+
 ## Rubric
 
 -	All required components are present and complete **Complete / Incomplete**