Instrument 3 Baseline equivalence template

INSTRUCTIONS
The information collected will be used for internal purposes to assess the equivalence of the treatment and
comparison samples.
1 At top of the sheet, indicate the analytic sample represented in the file.
Demographics
2 Choose whether you want to analyze age based on years (continuous) or grade level (categorical). This choice
should match how you plan to include age in your final analytic models. Any study enrolling youth in only one grade
should analyze age using years. YOU NEED ONLY REPORT ON AGE OR GRADE, NOT BOTH.
2a If you choose to analyze grade, construct the categories as you plan to analyze them in your final analytic models.
Enter those categories as the labels for Categories 1-6 and enter the number of individuals in each category in
Columns E and I. The percentages (automatically calculated in Columns C and G) should sum to 100 percent. If
you construct a variable with fewer rows than in the template, you should delete any empty rows. A chi-squared
statistic is calculated for this variable (provided there are no rows with zero totals).
3 Report the proportion of the sample that is female. Binary (%) outcomes should be entered as decimals in the
spreadsheet (e.g. 45% should be entered as 0.45).
4 Report the race and ethnic make-up of your sample. The four categories in Column C are the recommended
categories for your analysis, using a variable constructed from both race and ethnicity data. Enter the number of
youth in your sample each category in Columns E and I. The percentages automatically calculated in Columns C
and G should sum to 100 percent. A chi-squared statistic is calculated for this variable (provided there are no rows
with zero totals). NOTE: If you do not need some of the categories included in the template due to the makeup of
your sample, you can construct your own categories that better suit your study.
Behavioral measures
5 All grantees should report baseline equivalence on all core behavioral measures, already listed in the table.
Additionally, any other measures that are key outcomes for the evaluation should be added to the table and
assessed for equivalence between study groups.
6 When reporting on measures that youth only respond to on condition of having engaged in other activities (e.g.
pregnancy is only asked of youth who have ever engaged in vaginal sex), make sure to impute zero values for
those youth who were not asked to respond to the question because they had not engaged in the first activity. For
example, youth who have not ever had vaginal sex cannot have been pregnant or caused a pregnancy; you can
logically infer that those youth would have also responded "no" for pregnancy. This should be reflected in the data
file before calculating means or proportions of the conditional-response measures.
7 For any measures that are reporting frequency of an activity, use the TEMPLATE row included at the bottom of the
table. Copy this row for all variables measuring the number of times of an event to calculate equivalence of the
means.
Assessment
8 The spreadsheet will calculate p-values for each variable. This calculation assumes individual-level assignment to
condition. For RCTs or QEDs where assignment is at a level other than individual, evaluators must manually adjust
the p-values for clustering. These p-values should be entered into Column V.
9 Once all data is entered into the table, review the p-values calculated in Column U (or those clustered p-values in
Column V if appropriate), and assess equivalence of the groups.
10 This assessment should be conducted for all analytic samples of interest. For example, if the evaluation will look at
both 6-month and 12-month follow-up data, you will need to calculate and review baseline equivalence for those
two samples: those with 6-month follow-up data, and those with 12-month follow-up data. Create a copy the tab
with the table for each analytic sample, and enter the baseline values of the measures for the appropriate analytic
sample.

Treatment Group

Characteristics at BASELINE
Demographic characteristics
Age (in years)
Grade (counts)

[Category 1]
[Category 2]
[Category 3]
[Category 4]
Female (%)
Race (counts)
Non-Hispanic White
Non-Hispanic Black
Hispanic
Other race-ethnicities
Behavioral measures

Percentage
or
Unadjusted
Mean

Comparison Group

Standard
Deviation
(for
continuous Sample
variables)
Size

Percentage
or
Unadjusted
Mean

Group differences

Standard
Deviation
(for
continuous Sample
variables)
Size

0

0

0

0

t-statistic
df
(calculated (calculated
by the
by the
worksheet) worksheet)

[Core measure 1]

[Core measure 2]
[Core measure 3]

[Additional measure 1]

[Additional measure 2]
Notes: Please enter data in the yellow highlighted cells only. Please convert all yes/no responses to yes = one and no = zero in your datafile.

p-value
(calculated
by the
worksheet)

p-value adjusted for
clustering at level of
random assignment,
if applicable
(calculated by the
evaluator)