Hi Friends,

Even as I launch this today ( my 80th Birthday ), I realize that there is yet so much to say and do. There is just no time to look back, no time to wonder,"Will anyone read these pages?"

With regards,
Hemen Parekh
27 June 2013

Now as I approach my 90th birthday ( 27 June 2023 ) , I invite you to visit my Digital Avatar ( www.hemenparekh.ai ) – and continue chatting with me , even when I am no more here physically

Thursday, 20 March 2003

POPULATION PROFILES

20/08/03
Kartavya

Webservice / Population Profile

A candidate’s or an existing employee’s
Functional Exposure Profile

Can be compared with:

A. → Overall TOTAL POPULATION
(“Compare / Total” button)

OR

B. → Specific SUBSET POPULATION
(“Compare / Peers” button)


Under B, a person can be compared with other executives who are:
→ of same “Edu. Level”
→ of same “Exp. (yrs)”
→ of same “Design. Level”
→ of same “Industry”
→ of same “Age” etc.


For definition of “same Age” I suggest we take all persons born in:
→ preceding year
→ same year (as person being compared)
→ succeeding year

This means, if we want to compare a person born in 1972, then we take subset population of all persons born between
01-01-1971 up to 31-12-1973 (3 calendar yrs).

This becomes a specific “class-interval.” It will simplify matters considerably. Same person will appear in 3 subsets but it does not matter, since we are not adding-up “subsets” to arrive at “Total Population.”

Kartavya

Population Profiler

This will be one of the modules of our online web-service, “GLOBAL RECRUITER.”

We should give a catchy title to each module, e.g.:

  • Resummine

  • Re-Search (or shall we call it ResuSearch?)

  • Ad-Compose

  • India-Marketer (see my earlier note)

  • Population-Profiler
    etc. etc.


In respect of Population-Profiler, find enclosed:
→ 18 graphs
→ 1 User Interface


As usual, these are very preliminary & certainly not comprehensive. Please feel free to add / delete / modify. After a while, I will give you a concept-note on module SALARY-PROFILER.

As usual, this feature will be made available to ONLY those corporate subscribers who keep their resume database on OUR server.

Tariff:
→ Every “SHOW” click → $ 1.00
→ Every “PRINT” click → $ 2.00


Page (13) can produce 26 × 53 × 20 = 27,560 graphs!!
Ind. Func. Age Groups

Pattern-Recognition from Resumine Data

Frequency Distribution of “EDU. LEVEL”

(Sketch of bell curve: X-axis = Non-SSC, SSC, Diploma, Degree, PG, Doctoral; Y-axis = No. or %)
Population: _______
In: _______ (Month / Year)


If above graph was plotted, would we expect to see emergence of any pattern? If yes, what?

It would become apparent that:

  • X% of population is “Graduate”

  • Y% is “Post-Graduate” etc.

Shifts can be observed over the years. e.g., is “Entry-Level” shifting? One can also plot this graph “Ind-wise” / “Function-wise.”


If some definite pattern emerges, what logic / rules can we derive? What can such pattern “predict”?

  • A SSC cannot be in Finance Function or be a Finance Manager.

  • A Doctoral level is a must for R&D in Pharma.

  • You must have a Diploma to be a shop foreman.

  • A “MBA (Mktg)” is a must for Mktg Function.

One can review the logic, e.g.:

  • 95% of “Mktg” guys have MBA (Mktg).

  • 85% of “R&D” have M.Sc.

Pattern-Recognition from Resumine Data

Frequency Distribution of “BRANCH”

(Bar graph with X-axis: Pharmacy, Eng., Medical, Science, Arts, Komb. SC. Y-axis: No. or %)
Population: _______
In: _______ (Month / Year)


If above graph was plotted, would we expect to see emergence of any pattern? If yes, what?

With so many branches, we would need to plot bars in descending order (of no./%). & scroll horizontally.

This pattern may reveal the “demand” of executives belonging to diff. branches of education. It will tell students / jobseekers what “branch” to go for (assuming in real life, over a long period, demand/supply keep catching up / falling behind / overtaking each other).


If some definite pattern emerges, what Logic / Rules can we derive? What can such pattern “predict”?

This graph would be an excellent input to:
→ Education Planners / Edu. Institutes / Universities
→ Students / Parents
→ Jobseekers who wish to add “extra” qualifications

Pattern-Recognition from Resumine Data

Frequency Distribution of “Designation Levels”

(Bell curve with X-axis: Trainee, Sup, Off., Mgr., GM, VP, Pres., MD, CEO. Y-axis: No. or %)
Population: _______
In: _______ (Month / Year)

Note: This graph can also be prepared industry-wise.


If above graph was plotted, would we expect to see emergence of any pattern? If yes, what?

We expect this graph to “peak” around Manager level & then sharply taper off.

A company employing 30,000 persons can plot its own internal graph & superimpose to see if it follows UNIVERSAL pattern. If not, why? Is my company too “Top-Heavy”? Do I have too many “Generals” & too few “Foot-Soldiers”?

How does my curve compare with other companies in my industry?


If some definite pattern emerges, what Logic / Rules can we derive? What can such pattern “predict”?

This will be a powerful Manpower-Planning & Manpower Rationalization / Optimization tool!

Every HR manager would like to compare his own company curve with his industry-curve & make prompt/procedural recommendations to his corporate management. Over the years, he can “boast” how he has “altered” the curve thru SOUND manpower policies (read VRS).

Sign of a TOP-NOTCH HR manager! Corporate Mgrs would simply fall in love with this curve.

05-03-03
Pattern-Recognition from Resumine Database

Frequency Distribution of “TOTAL EXP (YRS)”
(Exp. goes up by 1 year on birth-date)

(Bell curve: X-axis = Total Experience (0–50 yrs), Y-axis = No. or %; Two lines plotted – Industry Norm vs Company A)
Population: _______
In: _______ (Month / Year)

Also plot:

  • Ind-wise

  • Func-wise


If above graph was plotted, would we expect to see emergence of any pattern? If yes, what?

This graph will follow same pattern as “AGE” graph. Any shift to the left (over the years) will show that executives are being “retired” earlier (than before).

New Economy Industries (IT / ITES / Telecom / Media / Enter) will find this graph skewed to left with more younger employees. Ind-wise & function-wise comparisons would unearth interesting patterns (within a company & across companies).


If some definite pattern emerges, what Logic / Rules can we derive? What can such pattern “predict”?

  • Is my company completely “out-of-sync” with my industry?

  • Am I having too many “greenhorns”?

  • Is my company population “aging” with high turnover of youngsters & NO turnover in above 15+ yr exp group?

  • Do I need to introduce VRS? For whom?

  • Do I need to hire 100 fresh graduate trainees? (Induction at bottom)

Pattern-Recognition from Resumine Database

CITY-WISE Frequency Distribution

(Bar chart: X-axis = Mumbai, Chennai, Delhi, Kolkata, Hyd, B’lore; Y-axis = No. or % with markers at 1.5L, 2L)
Population: _______
In: _______ (Month / Year)


If above graph was plotted, would we expect to see emergence of any pattern? If yes, what?

Vertical bars would need to be arranged in descending order – with a horizontal scroll-bar. Graph mainly shows “In which city is there maximum jobs?”

Of course, we could plot one separate graph for each city (function-wise / ind-wise / age-wise / exp-wise / education-wise) to show the demographic profile of professionals employed in any particular city.
E.g., PUNE city-profile would be of immense interest to HR Mgr. of Bajaj-Auto!


If some definite pattern emerges, what Logic / Rules can we derive? What can such pattern “predict”?

Plotted & compared over a long time, we may discover that “Hyderabad took over Bangalore in % of employment during last 3 years”!

By clicking on a bar (pertaining to a given city), we should be able to see the composition of professionals working in that city:

  • Industry-wise

  • Design-wise

  • Function-wise

  • Age-wise

  • Salary-wise

Pattern-Recognition from Resumine Data

“INDUSTRY-WISE” Frequency Distribution

(Bar chart: X-axis = IT, Eng, Chem, Pharma, FMCG, Telecom. Y-axis = No. or %)
Population: _______
In: _______ (Month / Year)


If above graph was plotted, would we expect to see emergence of any pattern? If yes, what?

Arranged in descending order (with horizontal scroll bar), this graph reveals sectoral employment.

Over a period, relative ranks would change as economy moves from “Manufacturing” to “Service” economy. “Sunrise” industries will overtake “Smoke-stack” industries.

To planners, this graph shows Employment Generation Potential of different industries.

By clicking on any Industry-Bar, one could dig deeper into profiles of professionals employed by that industry (Age / Exp / Edu / Designation / City-wise).


If some definite pattern emerges, what Logic / Rules can we derive? What can such pattern “predict”?

Pattern-Recognition from Resumine Database

AGE vs. EXPERIENCE (YRS) – Scattergram & Line of Best Fit

(Scatter plot: X-axis = Age (20–80 yrs), Y-axis = Exp (0–60 yrs). Line of Best Fit shown; anomalies marked “Impossible” and “Strange”)
Population: _______


If above graph was plotted, would we expect to see emergence of any pattern? If yes, what?

Age & Exp are very closely linked – in almost straight-line relationship. Of course, there will be a “BAND (Range) of Exp” for each age – but rather narrow.

Points falling far away from Line of Best Fit (Chi-square method?) are “Anomalies.”


If some definite pattern emerges, what Logic / Rules can we derive? What can such pattern “predict”?

The pattern can accurately predict:

  • Age (given total yrs. of experience) OR

  • Total Exp (given a person’s Age / DOB)

Any “anomalies,” while extracting data, can be immediately highlighted by software as ANOMALY!


Would you like me to now compile all 10 pages you’ve shared into a single, well-structured Word document — with clear section headings (Age, Edu Level, Branch, Designation Levels, Total Experience, City-wise, Industry-wise, Age vs Exp, etc.) and placeholders for graphs? This would give you a professional “Population Profiler & Pattern Recognition Report.”

Pattern-Recognition from Resumine Data

AGE vs. EDU. LEVEL
AGE vs. BRANCH

(Graph: X-axis = Age (20–80), Y-axis = Edu. Level / Branch; flat line shown at “Graduate” level)


If above graph was plotted, would we expect to see emergence of any pattern? If yes, what?

I do not expect any pattern beyond 20 yrs of Age (which, in any case, is our starting point!). Beyond that point, Edu. Level (rarely) changes with increasing Age, except for a very few fellows getting their late / part-time MBA!


If some definite pattern emerges, what Logic / Rules can we derive? What can such pattern “predict”?

Pattern-Recognition from Resumine Data

AGE vs. DESIGNATION LEVEL

(Graph: X-axis = Age (20–80), Y-axis = Designation Level: Trainee, Sup, Off, Mgr, GM, VP, Presi, MD. Line of Best Fit with Upper Band & Lower Band shown.)
Population: _______
In: _______


If above graph was plotted, would we expect to see emergence of any pattern? If yes, what?

By and large, a person goes up in “Designation Level” with increasing age – although not quite “linear” relationship.

Some people will climb hierarchy ladder very rapidly & some very slowly. Such persons falling above “upper-band” or falling below “lower-band” at any given age must be interviewed thoroughly! He could be working in “Daddy’s firm” or he could be a “nincompoop”!


If some definite pattern emerges, what Logic / Rules can we derive? What can such pattern “predict”?

“Age vs. Designation Level” graphs (along with Line of Best Fit / Upper Band / Lower Band) should be plotted for each Industry / each Function separately.

It would help a Recruitment Mgr in deciding what Designation to offer to a new employee.

A large company can plot such graphs of its employee population & use it (not to abuse it) while designating new recruits.

Pattern-Recognition from Resumine Data

EDU. LEVEL vs. DESIGNATION LEVEL

(Graph: X-axis = Designation Level – Tr, Sup, Off, Mgr, GM, VP, Presi, MD.
Y-axis = Edu. Level – Diploma, Graduate, P. Grad, Doctoral.
Line of Best Fit drawn through Graduate / Postgraduate levels.)


If above graph was plotted, would we expect to see emergence of any pattern? If yes, what?

There is no clear-cut relationship between the two, although % of employees having double degree / postgraduate edu. may be found to be higher amongst higher-designated employees.

My hypothesis is that “Edu. Level” is ONE of the (minor?) inputs while promoting a person to higher level.

E.g., Edu. Level is a very important input at ENTRY-LEVEL & important “Man-specs” in Job Adverts.


If some definite pattern emerges, what Logic / Rules can we derive? What can such pattern “predict”?

Pattern-Recognition from Resumine Data

Designation Level vs. Exp. (Yrs)

(Graph: X-axis = Designation Level – Tr, Sup, Off, Mgr, GM, VP, Presi, MD.
Y-axis = Exp. (Yrs), with Line of Best Fit, Upper Band, Lower Band.)


If above graph was plotted, would we expect to see emergence of any pattern? If yes, what?

Designation Level & Exp. (Yrs) have a direct (almost one-to-one) relationship.

Of course, persons falling outside the bands are “exceptions” to the rule. Some are FAST-TRACK & some are on SLOW-TRACK.

If a person is not even a Manager after 15 years of experience or a person is a VP after 9 years of experience; both are worth questioning thoroughly!


If some definite pattern emerges, what Logic / Rules can we derive? What can such pattern “predict”?

A company (employing large no. of executives at each designation level) may prepare such a graph of its own employees and then use it while recruiting executives from outside – so as not to upset the apple-cart!


Pattern Recognition
(13/07/03)

On a calendar that I have, there is a quote in Gujarati. Translated (loosely), it says:

“There have been learned persons in the past.
There are learned persons in the present.
There will be learned persons in future too.
They all followed same path.”

So, obviously, there is a definite “pattern.”

We could substitute “learned persons” by “successful executives.”

Then, we need to define,
“successful” and “path.”


In professional world of executives, broadly speaking, “success” is synonymous with:
→ Salary-level
→ Designation-level

Higher the salary and/or higher the designation-level, the more “successful” is the executive.

This paradigm/rule is universally accepted. This is then, THE TRUE DEFINITION of “success.”

This is the universal “Yardstick / Benchmark” for measuring “success.”

Of course, there are two important “modifiers / qualifiers” of this yardstick, viz:
▶ Age of executive (whose “success” is being measured)
▶ Company of executive (whose “success” is being measured)


This is obvious, as you can see from following examples:

#1
An executive is drawing a salary of Rs. 20 lakh at the age of 50, whereas another executive is drawing the same salary at the age of 30!

We would say that (obviously) the younger of the two is more successful than the older.


#2
An executive is “Vice-President” in a small (Rs. 50 cr. turnover) company, whereas another is “Vice-President” in a Rs. 500 cr. company (even if both are of same age).

Obviously, the Vice-President of bigger company is MORE SUCCESSFUL of the two!

So, to compute the SUCCESS INDEX of any given executive, we would need to plot following graphs for 100,000 (– or even a million) executives:

▶ A scattergram & a line of best fit will emerge.
▶ We could break up this overall total population into sub-populations of:
→ Each Industry (apple-for-apple comparison)
→ Each Function (apple-for-apple comparison)


Graph 1
(Designation Level vs Age)

We have data for the above & it should not be too difficult to plot this graph.

Next, we need to plot:

Graph 2
(Salary vs Age – with same remarks as above for “apple-for-apple comparisons”)

Here we have “Annual Salary” data of only those executives who have filled-in our online WEBFORM (resume). We could plot a graph for them.

These webforms will give you “Annual Salary” details (of different executives) for different “Years.”

e.g. Salary in the year:
1997–98
1998–99
1999–00
2000–01

(Sketch of a dropdown menu shown)

So, you will have a separate graph for each “year.”


So, if we have received 30,000 webforms in last 3 years, these may get broken up into 4/5 sub-populations (year-wise).

Don’t not bad for a beginning.


Real advantage would come, only (if and) when we return 70,000 email resumes to their owners (after duly converting), asking them to return after filling in:
→ Yearwise salary data
→ Designation (whenever changed)
(see my earlier notes)

They would take this trouble only if we promise (– and live up to our promise) of sending them graphs, showing where exactly they stand in comparison with overall population (– or his industry or his function population).


Once we have data about 100,000 executives, we are in a position to PREDICT!

We can say,

“If you are drawing this salary today (i.e., at your current age), you would most probably draw XYZ salary when you are of certain age.”
(of course, everything else remaining same)

OR

“If, at the age of 30, your current designation-level is MANAGER, you can (probabilistically) expect to be designated GENERAL MANAGER when you are 45.”
(once again, everything else remaining same)


Remember,

One does not have to be superstitious or believing in astrology, to get attracted to reading his Horoscope in newspapers daily – (since mostly it is so vaguely “reassuring”!).

You once said, www.salary.com has such a feature. If we can automate the content-compilation, we too can manipulate/process such content without human intervention.

(Two charts drawn: Designation Level vs Age; Salary vs Age, with multiple “Success Index” lines at 80, 90, 100, 110, 120, 130)


Depending upon the “spread” & “scatter” at each age-level, it should be possible to develop a set of LINES OF BEST FIT.

Mean/Median line (?) will represent 100% SUCCESS INDEX.

Then at each age level, we can plot frequency distribution & compute σ (Std. deviation).

Then:

  • Central line = 100% Success Index

  • +1σ = 110%

  • +2σ = 120%

  • +3σ = 130%

  • –1σ = 90%

  • –2σ = 80%

  • –3σ = 70%



Kartavya
06-03-03

Population Profiler

This will be one of the modules of our online web-service, “GLOBAL RECRUITER.”

We should give a catchy title to each module, e.g.

  • Resumine

  • Re-Search (or shall we call it ResuSearch?)

  • Ad-Compose

  • India-Marketer (see my earlier note)

  • Population-Profiler

  • etc. etc.


In respect of Population-Profiler, find enclosed:
→ 13 graphs
→ 1 User Interface


As usual, these are very preliminary & certainly not comprehensive. Please feel free to add/delete/modify. After a while, I will give you a concept note on module SALARY-PROFILER.

As usual, this feature will be made available to ONLY those corporate subscribers who keep their resume database on OUR server.


Tariff:
→ Every “SHOW” click → $1.00
→ Every “PRINT” click → $2.00


Page 13 can produce 26 × 53 × 20 = 27,560 graphs!!
(Ind / Func / Age Groups)

05-03-03
Pattern-Recognition from Resumine Database

Frequency Distribution of “AGE”

(Bell curve shown, X-axis = Age 20–80, Y-axis = No. or %)
Population: _______
In: _______ (Month / Year)


If above graph was plotted, would we expect to see emergence of any pattern? If yes, what?

It will tell us no. of executives in each Age-Group (Class-Interval). As years go by & population size grows, we will see definite shifting of the mean/average/median, to the right or to the left.

Standard deviations (σ) can be computed & displayed on each graph. June 2000 & June 2003 graphs can be superimposed (different colors) to see the “shift.”


If some definite pattern emerges, what Logic/Rules can we derive? What can such pattern “predict”?

After observing the “shift” over a few years, we would be able to “predict” the distribution after, say, 10 years.


Note:
One year to be added to each person’s “Age” on his birthday to keep data up-to-date.

05-03-03
Pattern-Recognition from Resumine Database

Frequency Distribution of “EDU. LEVEL”

(Bell curve shown; X-axis = Non SSC, SSC, Diploma, Degree, PG, Doctoral; Y-axis = No. or %)
Population: _______
In: _______ (Month / Year)


If above graph was plotted, would we expect to see emergence of any pattern? If yes, what?

It would become apparent that:

  • X% of population is “Graduate”

  • Y% is “Post-Graduate” etc.

Shifts can be observed over the years, e.g. Is “Entry-Level” shifting? One can also plot this graph “Ind-wise” / “Function-wise.”


If some definite pattern emerges, what Logic/Rules can we derive? What can such pattern “Predict”?

  • A SSC cannot be in Finance Function or be a Finance Manager.

  • A Doctoral level is a must for R&D in Pharma.

  • You must have a Diploma to be a Shop Foreman.

  • A “MBA (Mktg)” is a must for “Mktg” function.

(One can reverse the logic, e.g.:

  • 95% of “Mktg” guys have MBA (Mktg) (Predict).

  • 85% of “R&D” → M.Sc. (” ”).

05-03-03
Pattern-Recognition from Resumine Database

Frequency Distribution of “BRANCH”

(Bar chart shown; X-axis = Pharmacy, Eng., Medical, Science, Arts, Komb. Sc.; Y-axis = No. or %)
Population: _______
In: _______ (Month / Year)


If above graph was plotted, would we expect to see emergence of any pattern? If yes, what?

With so many branches, we would need to plot bars in descending order (of no./%). & scroll horizontally. This pattern may reveal the “demand” of executives belonging to different branches of education.

It will tell students/jobseekers what “branch” to go for.
(Assuming, in real life, over a long period, demand/supply keep catching up/falling behind/overtaking each other)


If some definite pattern emerges, what Logic/Rules can we derive? What can such pattern “Predict”?

This graph would be an excellent input to:
→ Education Planners / Edu. Institutes / Universities
→ Students / Parents
→ Jobseekers who wish to add “extra” qualifications

05-03-03
Pattern-Recognition from Resumine Database

Frequency Distribution of “Designation-Levels”

(Bell curve shown; X-axis = Trainee, Sup, Off, Mgr, GM, VP, Pres, MD, CEO; Y-axis = No. or %)
Population: _______
In: _______ (Month / Year)
(Note: This graph can also be prepared Industry-wise)


If above graph was plotted, would we expect to see emergence of any pattern? If yes, what?

We expect this graph to “peak” around “Manager” level & then sharply taper off.
A company employing 30,000 persons can plot its own internal graph & superimpose, to see if it follows Universal Pattern.
If not, why?

  • Is my company too “Top-Heavy”?

  • Do I have too many “Generals” & too few “Foot-Soldiers”?

  • How does my Curve compare with other companies in my Industry?

If some definite pattern emerges, what Logic/Rules can we derive? What can such pattern “Predict”?

This will be a powerful Manpower-Planning & Manpower Rationalization / Optimization Tool.
Every HR manager would like to compare his own Company Curve with his Industry-curve & make formal/procedural recommendations to his Corporate Management.

Over the years, he can “boast” how he has “altered” the Curve thru Sound manpower policies (read, VRS)!!

A sign of a Top-Notch HR manager! Corporate Mgmt. would simply fall in love with this Curve.

05-03-03
Pattern-Recognition from Resumine Database

Frequency Distribution of “Total Exp (Yrs)”
(Exp. goes up by 1 year on birth-date)

(Bell curve shown; X-axis = 0–50 years; Y-axis = No. or %)
Two lines: “Industry Norm” (bell curve) & “Company A” (flatter, skewed).
Population: _______
In: _______ (Month / Year)

Also plot:

  • Ind-wise

  • Func-wise

If above graph was plotted, would we expect to see emergence of any pattern? If yes, what?

This graph will follow same pattern as “AGE” graph.

Any shift to the left (over the years) will show that executives are being “Retired” earlier (than before).

New Economy Industries (IT/ITES/Telecom/Media/Enter) will find this graph “skewed” to left with more younger employees.

Ind-wise & function-wise comparisons would unearth interesting patterns (within a company & across companies).


If some definite pattern emerges, what Logic/Rules can we derive? What can such pattern “Predict”?

  • Is my company completely “out-of-sync” with my Industry?

  • Am I having too many “Green-Horns”?

  • Is my company population “Aging” with high turnover of youngsters & no turnover in above 15-yr exp. group?

  • Do I need to introduce VRS? For whom?

  • Do I need to hire 100 fresh graduate trainees? (Induction at bottom).


05-03-03
Pattern-Recognition from Resumine Database

City-Wise Frequency Distribution

(Bar chart drawn; X-axis = Mumbai, Chennai, Delhi, Kolkata, Hyd, B’lore; Y-axis = No. or % with 2 Lakh and 1.5 Lakh noted on scale)
Population: _______
In: _______ (Month / Year)


If above graph was plotted, would we expect to see emergence of any pattern? If yes, what?

Vertical bars would need to be arranged in descending order – with a horizontal scroll-bar.
Graph merely shows “In which city is there max. jobs?”

Of course, we could plot one separate graph for each city (function-wise/ind-wise/age-wise/exp-wise/edu-wise etc.) to show the Demographic Profile of professionals employed in any particular city.

E.g. Pune city-profile would be of immense interest to HR Mgr. of Bajaj Auto!


If some definite pattern emerges, what Logic/Rules can we derive? What can such pattern “Predict”?

Plotted & compared over a long time, we may discover that “Hyderabad took over Bangalore in % employed during last 3 years”!

By clicking on a bar (pertaining to a given city), we should be able to see the composition of professionals working in that city:

  • Industry-wise

  • Edu-wise

  • Age-wise

  • Function-wise

  • Design-wise

  • Salary-wise


Would you like me to merge these transcriptions with the earlier ones into a single structured Word file, arranged graph-by-graph (Age, Edu Level, Branch, Designation, Exp, Industry, City, etc.), so you have a consolidated “Population Profiler” draft?


05-03-03
Pattern-Recognition from Resumine Database

Age vs. Experience (Yrs) – Scattergram & Line of Best Fit

(Graph drawn: X-axis = Age (20–80), Y-axis = Exp (0–60 yrs). Diagonal line of best fit, with notes “Impossible!” at 50 yrs Exp vs 30 yrs Age, and “Strange!” near 70 yrs Age. Population blank.)


If above graph was plotted, would we expect to see emergence of any pattern? If yes, what?

Age & Exp. are very closely linked – in almost straight-line relationship.
Of course, there will be a “band” (range) of experience for each age – but rather narrow.

Points falling far away from “Line of Best Fit” (Chi-square method?) are “Anomalies.”


If some definite pattern emerges, what Logic/Rules can we derive? What can such pattern “Predict”?

The pattern can accurately predict:

  • Age (given total yrs of experience)
    OR

  • Total Exp. (given a person’s Age/DOB)

Any “anomalies,” while extracting data, can be immediately “highlighted” by software.
ANOMALY!

05-03-03
Pattern-Recognition from Resumine Database

Age vs. Edu. Level
Age vs. Branch

(Graph drawn: X-axis = Age (20–80), Y-axis = Edu Level / Branch. Horizontal line labeled “Graduate” at ~20 yrs age onward.)


If above graph was plotted, would we expect to see emergence of any pattern? If yes, what?

I do not expect any pattern beyond 20 yrs of Age (which, in any case, is our starting point!).
Beyond that point, Edu. Level rarely changes with increasing Age – except for a very few fellows getting their late/part-time MBA!


If some definite pattern emerges, what Logic/Rules can we derive? What can such pattern “Predict”?
(No additional notes here, left blank.)


Would you like me to now consolidate all transcribed pages (1–10/13 so far) into a structured digital draft (Word/PDF), preserving diagrams as placeholders for digital charts, so you have one complete Population Profiler Concept Note?

05-03-03
Pattern-Recognition from Resumine Database

Age vs. Designation-Level

(Graph drawn: X-axis = Age (20–80), Y-axis = Designation Level (Trainee, Sup, Off, Mgkr, GM, VP, Presi, MD). A “Line of Best Fit” runs diagonally, with “Upper Band” and “Lower Band” shown.)


If above graph was plotted, would we expect to see emergence of any pattern? If yes, what?

By and large, a person goes up in “Designation-Level” with increasing age – although not quite “linear” relationship. Some people will climb hierarchy-ladder very rapidly & some very slowly.

Such persons, falling above “Upper-Band” or falling below “Lower-Band” at any given age, must be interviewed thoroughly!
He could be working in “Daddy’s firm” or he could be a “Nincompoop”!


If some definite pattern emerges, what Logic/Rules can we derive? What can such pattern “Predict”?

“Age vs. Designation Level” graphs (along with Line of Best Fit / Upper Band / Lower Band) should be plotted for each “Industry” / each “Function” separately.

It would help a Recruitment Manager in deciding what “Designation” to offer to a candidate.

A large company can plot such graphs for its own employee-population & use it, not to upset apple cart while designating “new-recruits.”

05-03-03
Pattern-Recognition from Resumine Database

Edu. Level vs. Designation-Level

(Graph drawn: X-axis = Designation Level (Trainee → MD), Y-axis = Edu Level (Diploma, Graduate, P. Grad, Doctoral). A “Line of Best Fit” rises stepwise.)


If above graph was plotted, would we expect to see emergence of any pattern? If yes, what?

There is no clear-cut relationship between the two, although % of employees having double degree / post-graduate education may be found to be higher amongst higher-designated employees.

My hypothesis is that “Edu Level” is one of the (minor?) inputs while promoting a person to higher level.

But “Edu Level” is a very important input at Entry Level & important “Man-Specs” in Job Adverts.


If some definite pattern emerges, what Logic/Rules can we derive? What can such pattern “Predict”?
(No further notes, space left blank.)

05-03-03
Pattern-Recognition from Resumine Database

Designation-Level vs. Exp. (Yrs)

(Graph drawn: X-axis = Designation Level (Trainee → MD), Y-axis = Exp (0–50 yrs). “Line of Best Fit” shown with “Upper Band” and “Lower Band”.)


If above graph was plotted, would we expect to see emergence of any pattern? If yes, what?

Designation-Level & Exp. (Yrs) have a direct (almost one-to-one) relationship.
Of course, persons falling outside the bands are “exceptions” to this rule. Some are FAST-TRACK & some are on SLOW-TRACK.

If a person is not even a “Manager” after 25 years of experience or a person is a VP after 8 years of experience, both are worth questioning thoroughly!


If some definite pattern emerges, what Logic/Rules can we derive? What can such pattern “Predict”?

A Company (employing large no. of executives at each designation-level) may prepare such a graph of its own employees and then use it while recruiting executives from outside – so as not to upset the apple-cart!


Page 14 (13/13)

Frequency Distribution
Industry: AGE (Yrs)

(Graphs drawn: multiple bell-curves at different ages – 25, 30, 35, 40, 45 – plotted against Designation levels (Tr, Sup, Off, Mgr, GM, VP, Pres). Population blank for each age.)


Notes:
Such graphs can be plotted:

  • Industry-wise (26 Industries)

  • Function-wise (53 Functions)

  • Total population-wise






































No comments:

Post a Comment