In general the PTC measures graduation in three different ways:
- On an absolute time scale
- On a relative time scale, which itself is measured in two ways:
- Cumulatively
- Cross sectionally
- Without regard to time scale
Measures of graduation that are on an absolute time scale indicate the semester a student graduated in terms of a calendar year (e.g., September 1st, 2008). [In other words, a student with graduation date of 9/1/2008 is a student who graduated at the end of the fall 2008 term]
For the majority of analyses, relative-scale, cumulative graduation measures are preferable. These measures indicate, at a given point of time relative to entry to college, whether or not a student has graduated at some point between entry and that point in time. That is, a four-year graduation measure will indicate if a student earned a degree within four years from graduation. For that same student, a six-year cumulative graduation measure will also completion in that time period, even though the time it took the student to graduate was shorter than six years.
Relative-scale, cross sectional measures of graduation indicate whether or not a student graduated at a particular point in time relative to their point of entry to CUNY. Thus for a student who graduates in eight semesters, their value for graduating in semester eight will indicate completion but their value in semester twelve will not indicate completion. These measure have been used for sequence analysis in the past.
The measures of graduation that do not involve a time scale indicate whether or not a student ever got a degree within the time frame of the data set.
It should be noted that the graduation rates produced by the PTC will not match those published on OIRA's
website. This is due to
differences in the way that the PTC measures entering cohorts and is to be expected.
The absolute time scale measures in the PTC include:
- Date a degree was received in string format (M/D/YYYY) [based on term_graduated_date in degree facts]
- Date a degree was received in Stata date format
- cert_deg_date
- aa_deg_date
- ba_deg_date
- Note that the absolute time scale variables measure the earliest instance of a given degree type for a given student. So if a student earns an associate degree and then comes back and earns a second associate degree, their value for aa_date will be the date of the ealier degree. This does not affect students who earn multiple types of degrees.
The relative-scale, cumulative graduation measures in the PTC include:
- System measures of graduation:
- These indicate whether or not a student earned a degree from any CUNY institution
- These measures include
- Baccalaureate degree completion
- ba_100, ba150, ba_200
- These measures and all others that end in _100 - _400 indicated completion within a certain percent of the expected rate. For baccalaureate degrees, the expected rate is four years. For associate degrees, the expected rate is two years. This means that ba_100 and aa_200 both indicate whether or not a student earned a degree within four years of entering CUNY
- Associate degree completion
- aa_100, aa_150, aa_200, aa_300, aa_400
- Combined degree completion
- cdeg01 - cdeg10
- These are measured at the yearly level and include the following values
- 1 - Certificate
- 2 - Associate
- 3 - Baccalaureate
- There are recoded versions of these variables (r_cdeg01 - r_cdeg10) where missing values are recoded to zero
- These variables measure the highest degree earned at a given point relative to entry. If a student earns an associate in their second year and a baccalaureate in their fourth year, they will have a value of 2 for cdeg02 and cdeg03 and a value of 3 for cdeg04 - cdeg10.
- The degree specific variables above are constructed from the combined degree completion variables in Stata using the following syntax:
- gen aa_100=cdeg02==2
gen aa_150=cdeg03==2
gen aa_200=cdeg04==2
gen aa_300=cdeg06==2
gen aa_400=cdeg08==2
gen ba_100=cdeg04==3
gen ba_150=cdeg06==3
gen ba_200=cdeg08==3 - Certificate versions of any of these variables could be constructed if needed for an analysis by conditioning on cdeg** == 1. The expected time to degree for a certificate is one year. Note that these certificates are not the same as advanced certificates
- Retention
- While not solely a measure of degree completion, the year-based retention variables (ret01 - ret10) include completion.
- Time to degree
- time_to_grad_cert, time_to_grad_aa, time_to_grad_ba
- These variable measures the # of days from entry to receiving a degree using the following syntax:
- gen time_to_grad_cert = cert_deg_date - entry_date
gen time_to_grad_aa = aa_deg_date - entry_date
gen time_to_grad_ba = ba_deg_date - entry_date - Note: Because the degree date is tied to the term in which the student graduated, this calculation will under-report time to degree by at least 3 months. For example, a student who earned their degree after the Spring 2008 term will have a degree date of 2/1/2008, although the student did not actually graduate until the end of that term (e.g., 5/31/2018). Use carefully and adjust accordingly.
- years_to_cert, years_to_aa, years_to_ba
- gen years_to_cert = (cert_deg_date - entry_date)/365.25 if cert_deg_date!=.
- gen years_to_aa = (aa_deg_date - entry_date)/365.25 if aa_deg_date!=.
- gen years_to_ba = (ba_deg_date - entry_date)/365.25 if ba_deg_date!=.
- Note: see previous not for time_to_grad variable.
- Institutional measures of graduation
- These indicate whether or not a student earned a degree from their first CUNY institution (e.g. if a student starts at BMCC, transfers to Baruch without earning an associate degree, then earns a baccalaureate degree at Baruch, they will not be counted as a graduate in these measures)
- These measures include
- Baccalaureate degree completion
- ba_inst_100, ba_inst_150, ba_inst_200
- Associate degree completion
- aa_inst_100, aa_inst_150, aa_inst_200, aa_inst_300, aa_inst_400
- Combined degree completion
- grad_inst_yr01- grad_inst_yr10
- These variables are constructed using the following syntax
The cross sectional relative-scale measures include:
- grad_sem02 - grad_sem20
- They are constructed using in Stata the following syntax
- These measures are measure graduation within the entire CUNY system and are not limited to earning a degree from the same school that a student entered CUNY
The measures of graduation that do not include a time scale are:
- anyba, anyaa, anycert
- These variables measure whether or not a student ever received a degree of a certain type within the time frame of the data set (i.e. up to 10 years).
- There are recoded versions of the first two (r_anyba, r_anyba) where missingness was recoded to zero.