Compare Alberta Bartender Wage Floors in Job Bank Releases
Compare bartender wage floors in Alberta’s September and October Job Bank releases. Read hourly pay alongside posting counts, vacancies and regional coverage.

The Canada Job Bank Posting and Hourly Wage Trends actor compares the hourly minimum rates employers advertised in two monthly releases. For Edmonton bartenders, the mean rose from CAD 15.00 in September 2026 to CAD 15.23 in October. Fewer posting-location records contributed in October, yet those records reported more vacancies. A regional workforce bulletin can report all three movements together. The wage figure describes the advertisements captured by Job Bank. A paycheque or a completed hire would require different evidence.
Why read Edmonton’s wage and vacancy figures together?
A hospitality association preparing an Alberta bulletin might hear that Edmonton bars were looking for more bartenders in October. Is that visible in the advertised wages, the number of postings, or the positions those postings represented? Each measure moved differently. The September file contained 13 Edmonton posting-location records for this occupation, all with hourly minimums. October contained eight. Between them, the eight October records reported 32 vacancies, up from 21 the month before. September Job Bank release · October Job Bank release
The Alberta comparison returned eight groups for the same occupation, including one whose region Job Bank did not specify. Calgary’s records tell a different story. Its mean advertised hourly minimum went from CAD 16.11 across nine records to CAD 17.05 across eleven. Knowing the denominator lets an editor describe the change precisely and ask whether a second month of data supports a longer-lasting pattern. Canada Job Bank actor and dataset fields
Which source fits the wage question?
The word wage could refer to an offer in an advertisement, an employee’s earnings or a wider statistic for an occupation. Decide what your table will claim before choosing its source. A dated observation of an offer can support a historical bulletin even if the advertisement has since closed.
| Method | Question it can answer | What to carry into the report |
|---|---|---|
| Job Bank monthly files, grouped by this actor | What minimum hourly rates and vacancy counts did Job Bank postings disclose for this occupation and region? | Both release months, the reporting denominators and the named region |
| Job Bank monthly files, analysed directly | Which individual records or other published fields explain the composition of a regional result? | The source snapshot ID and explicit rules for missing pay, time units and aggregation |
| Statistics Canada’s Job Vacancy and Wage Survey | What do survey estimates say about job vacancies and offered wages across a wider employer population? | The survey’s reference period, geography, definitions and published quality guidance |
Statistics Canada publishes detailed quarterly vacancy information by occupation and economic region. Use that series when the question calls for a survey estimate. A mean calculated from the hourly floors disclosed in Job Bank advertisements has a different reporting population and a different definition. Statistics Canada survey methodology
How do I compare two Job Bank releases for the same occupation?
The following Python blocks run in order. They use the apify-client package and an Apify API token held in the APIFY_TOKEN environment variable. You can also enter the same three input values in the actor form. Keep the month fixed if you plan to reproduce the figures in a report.
Which occupation and dates belong in the input?
NOC 2021 assigns 64301 to bartenders, including bartender supervisors. It lists bar managers and food and beverage servers separately. Use the five-digit code as text even when a code starts with zero. The input below selects Alberta and ends in October 2026. The actor automatically pairs October with September. A query for a word such as “bar” would cover a different set of occupations. Official NOC 64301 occupational profile
import os
from apify_client import ApifyClient
client = ApifyClient(os.environ["APIFY_TOKEN"])
query = {"endMonth": "2026-10", "province": "Alberta", "nocCode": "64301"}How do I keep the result attached to the right releases?
Call the actor once and read the dataset for that run. The region comes from the source, so a row can legitimately have an empty region. Keep it when checking provincial totals. The worked comparison includes a single October record in that group, with one reported vacancy and a disclosed hourly minimum of CAD 20.51. Assigning that record to Calgary would change Calgary’s results without evidence for the location.
run = client.actor("thirdwatch/canada-job-bank-trends").call(run_input=query)
if not run or run["status"] != "SUCCEEDED":
raise RuntimeError("The Job Bank release comparison did not complete")
rows = list(client.dataset(run["defaultDatasetId"]).iterate_items())
if not rows:
raise ValueError("No records matched the occupation and province in either release")
by_region = {row["economicRegion"]: row for row in rows}
assert all(row["scope"] == "full-monthly-releases" for row in rows)The row’s month and previousMonth identify the advertising periods. resourceModified and previousResourceModified record when Open Canada says it last changed the two resources. The October file used here shows a modification date of November 4, 2026. Date your chart by the advertising month and retain the modification date and file URL in a source note. Job Bank release catalogue
How much hourly pay did the records actually disclose?
Use recordsReportingHourlyMinimum as the denominator for the mean. A record must disclose an hourly pay period and a minimum amount to contribute. Each qualifying location record has equal weight, regardless of how many vacancies it advertises. Compare that denominator with postingLocationRecords for both periods. A movement in disclosure coverage can change which advertisements supply your wage figure.
for region in ("Calgary Region", "Edmonton Region"):
row = by_region[region]
for period in ("previous", "current"):
m = row[period]
print(region, row["previousMonth" if period == "previous" else "month"],
m["recordsReportingHourlyMinimum"], m["postingLocationRecords"],
m["meanAdvertisedHourlyMinimumCAD"])In these two regions every selected record supplied an hourly minimum in both months. The term minimum refers to the lower end of each advertisement’s pay range. Calgary’s October figure of CAD 17.05 is the arithmetic mean of eleven advertised floors. Finding the single cheapest advertised rate would require looking back at those eleven amounts individually. A salary midpoint, a median and a vacancy-weighted rate would also each require their own calculation.
What can I put in a bulletin when hourly data is missing?
Print a wage movement only when the source supports it in both periods. Keep a known zero count distinct from an unavailable rate. Red Deer had no matching posting-location record in September and two in October, with a mean October floor of CAD 15.00. The October value can appear in a table on its own. The monthly wage change belongs in the same missing-data category as any other comparison without two disclosed means.
for region, row in by_region.items():
change = row["change"]["meanAdvertisedHourlyMinimumCAD"]
label = "insufficient hourly observations" if change is None else f"{change:+.2f} CAD/hour"
print(region or "Region unspecified", label)What do the September and October sample rows show?
This projection comes from the Edmonton row of the complete two-release comparison. The shorter example preserves all three counts you need to check the bulletin’s wording.
{
"economicRegion": "Edmonton Region",
"previousMonth": "2026-09",
"month": "2026-10",
"previous": {"postingLocationRecords": 13, "reportedVacancies": 21, "recordsReportingVacancies": 13, "meanAdvertisedHourlyMinimumCAD": 15, "recordsReportingHourlyMinimum": 13},
"current": {"postingLocationRecords": 8, "reportedVacancies": 32, "recordsReportingVacancies": 8, "meanAdvertisedHourlyMinimumCAD": 15.23, "recordsReportingHourlyMinimum": 8},
"change": {"postingLocationRecords": -5, "reportedVacancies": 11, "meanAdvertisedHourlyMinimumCAD": 0.23}
}The records with disclosed vacancy counts numbered 13 and eight, respectively. That makes the rise of eleven reported vacancies directly interpretable in this case. For a group where only part of the file supplied vacancy counts, describe the sum as reported vacancies and show the number of contributing records. If a group had postings but none supplied any vacancy count, the sum stays blank. A group with no postings in a month has a count of zero.
The Wood Buffalo–Cold Lake Region illustrates a different editorial choice. Its advertised mean fell from CAD 19.42 to CAD 15.00, a much larger change than Edmonton’s, but October contributed a single hourly observation against four in September. Give the counts as much prominence as the difference. A reader can decide what to make of one advertisement without assuming it establishes the going rate across the region.
Which pitfalls change the story in an Alberta workforce report?
Repeated appearances can inflate an apparent hiring total. Job Bank’s underlying identifier is a posting-location snapshot ID. An advertisement can cover multiple locations, and a monthly comparison does not follow a worker, a firm or a hiring outcome. Keep a label such as “advertised posting-location records” on a count chart. Adding consecutive months together produces a sum of monthly observations. Establish a separate matching method if your question concerns unique advertisements. Job Bank release data dictionary
Regional averages also need care. The eight groups together contained 28 posting-location records and 37 reported vacancies in September, then 27 records and 52 vacancies in October. Every record in this example disclosed a vacancy count. For wages, averaging the available regional means gives a region with one hourly observation the same influence as Calgary’s eleven in October. An exact provincial wage mean requires adding the underlying qualifying amounts and dividing by their total reporting count. Reweighting the displayed means can only approximate that result because the actor rounds them to cents.
Attribution is part of a usable export. Identify Employment and Social Development Canada, Job Bank as the source. Keep the two resource URLs and identify the calculation as your regional summary. Link the Open Government Licence – Canada alongside the chart or dataset wherever possible. This makes it easier for the next reader to find the historical files, check the figures and understand who produced the analysis. Open Government Licence – Canada
Related use cases
Searching advertised salary bands serves a different buyer question. For individual roles that meet pay criteria, see the remote engineering salary-band guide. To design a series using named technology fields, the remote hiring trends guide shows how a posting-based market signal depends on the fields that source contains.
If the hiring team needs experience-specific comparisons before it sets a range, the experience-based salary benchmark guide covers that task using a different source and market. The job-board data guide and Thirdwatch article index cover additional ways to work with advertisements.
Frequently asked questions
Does a higher October wage floor mean individual bartenders got a raise?
The figures summarize minimum hourly rates in two sets of advertised posting-location records. Employers or locations can differ from one month to the next. A claim about a worker’s raise needs evidence of that person’s pay across the same period.
Why does Edmonton show fewer posting records and more vacancies?
One posting-location record can advertise several vacancies. Edmonton’s 13 September records reported 21 vacancies, while its eight October records reported 32. All the records in both groups supplied a vacancy count. A separate source of hiring outcomes would be needed to establish how many positions employers subsequently filled.
What should a blank month-to-month wage change look like in a chart?
Leave a gap and label which month lacks a qualifying hourly minimum. Camrose–Drumheller had one September observation and no October record in this occupation. A zero plotted at October would invent an hourly rate and create a spurious decline.
Can I use a pair of excerpts to test my dashboard?
Two adjacent official excerpts can exercise the same fields and null handling. Supply the month, official resource URL and CSV or tab-separated text for each excerpt. Keep the resulting supplied-excerpts scope in the dashboard so colleagues can see that totals cover the supplied rows.