0819_ss_PartB_020411

3060-0819

February 2011

B. Collections of Information Employing Statistical Methods:

1. This collection employs a statistical method in order to reduce the public burden. The universe of respondents is ETCs receiving Lifeline/Link Up support in federal default states.

The Lifeline Order requires certain ETCs to verify annually that a statistically valid sample of their Lifeline recipients receiving support continue to be eligible under the federal eligibility criteria. The Lifeline Order’s requirement applies only to those ETCs with Lifeline customers from federal default states. A federal default state is a state or territory that either (1) has adopted the federal eligibility criteria for Lifeline/Link Up, or (2) does not have its own state-based Lifeline/Link Up program.

Federal default states and territories include: American Samoa, Delaware, Hawaii, Iowa, Indiana, Louisiana, New Hampshire, North Dakota, Northern Mariana Islands, and South Dakota. There are currently 388 ETCs in these states and territories. The total number of subscribers in these states and territories is approximately 250,000. The Commission anticipates that the number of ETCs and subscribers will continue to increase over time.

Because the Lifeline Order requires ETCs to submit the results of their sampling, we estimate that the response rate will exceed 95%.

2. Each year, ETCs in federal default states are required to verify the continued eligibility of a statistically valid sample of their Lifeline customers. The size of the statistically valid sample varies based on the number of Lifeline subscribers (N) and the previously estimated proportion of Lifeline subscribers inappropriately taking Lifeline service (P).

For ETCs with 400,000 Lifeline subscribers or less, the appropriate sample size is calculated pursuant to the following formula: sample size = N/(1+{[N-1]/n}). N is the number of Lifeline subscribers and n = 2.706 * P*(1 – P) / .000625, where P is the previously estimated proportion of Lifeline subscribers inappropriately taking Lifeline service. We note that the values 2.706 and .000625 in this formula are mandated by OMB.

ETCs should use the results of samples from previous years to determine the estimated proportion (P) of subscribers inappropriately taking Lifeline service. In all instances, the estimated proportion (P) should never be less than .01 or more than .06.

To simplify the calculation for respondents, the Lifeline Order provides a table of sample sizes created using this formula. The table uses two variables N (number of Lifeline subscribers) and P (previously estimated proportion of Lifeline subscribers inappropriately taking Lifeline service). To determine the correct sample size, respondents select the box that matches its number of Lifeline subscribers N and the proportion P. If the exact values for N and P are not listed on the table, ETCs are directed to round up to the nearest value on the table. The table is attached for clarity.

For example, if an ETC found in the prior year that 3.8 percent of its 9,500 Lifeline subscribers inappropriately took Lifeline service, the ETC would use a sample size of 164 (value using 10,000 customers and proportion .04). The ETC in this example would be required to sample 164 subscribers to verify the subscribers’ continued eligibility under the federal eligibility criteria.

Because the adjustment for the number of Lifeline subscribers is de minimis above 400,000 subscribers, ETCs with more than 400,000 Lifeline subscribers must use a specific formula to calculate the sample size. For ETCs with more than 400,000 Lifeline subscribers, the ETC will determine a statistically valid sample size using the following formula:

Sample Size = 2.706 * P*(1 – P) / .000625.

P is the previously estimated proportion of Lifeline subscribers inappropriately taking Lifeline service.

3. ETCs in federal default states with Lifeline subscribers are required by the Commission’s rules to submit this information annually to USAC. The Administrator of the Universal Service mechanism will be able to determine which ETCs have not submitted the required information and will follow up by letter with those who have not responded.

4. The Administrator provides training information on how to calculate the sample on its website. The sampling formula relies on the prior year’s actual results. Use of the prior year’s results will further the Commission’s goal of making sure that only eligible subscribers receive support.

5. Contact names and phone numbers. For the statistical aspects of the design, the contact is

Suzanne Mendez, Wireline Competition Bureau, Federal Communications Commission, [email protected]. Her telephone number is (202) 418-0940.

For the collection of the completed samples, the contact is Pamela Gallant, Director, Low Income Programs, Universal Service Administrative Company, [email protected]. Her telephone number is (202) 776-0200.

Sample Size Table

Previously Estimated Proportion of Subscribers Inappropriately Taking Lifeline Service (P)¹

(N) Number of Lifeline Subscribers	0.01	0.015	0.02	0.025	0.03	0.035	0.04	0.045	0.05	0.055	0.06
400,000	43	64	85	106	126	146	166	186	206	225	244
100,0002	43	64	85	105	126	146	166	186	206	225	244
90,000	43	64	85	105	126	146	166	186	205	224	244
70,000	43	64	85	105	126	146	166	186	205	224	243
60,000	43	64	85	105	126	146	166	185	205	224	243
30,000	43	64	85	105	125	146	165	185	204	223	242
20,000	43	64	85	105	125	145	165	184	204	223	241
15,000	43	64	84	105	125	145	164	184	203	222	240
10,000	43	64	84	104	124	144	164	183	202	220	238
9,000	43	64	84	104	124	144	163	182	201	220	238
8,000	43	63	84	104	124	144	163	182	201	219	237
7,000	43	63	84	104	124	143	162	181	200	218	236
6,000	43	63	84	104	123	143	162	180	199	217	235
5,000	43	63	83	103	123	142	161	179	198	215	233
4,000	42	63	83	103	122	141	160	178	196	213	230
3,000	42	63	83	102	121	139	158	175	193	209	226
2,000	42	62	81	100	119	136	154	170	187	202	218

Sample Size Table

Previously Estimated Proportion of Subscribers Inappropriately Taking Lifeline Service (P)

(N) Number of Lifeline Subscribers	0.01	0.015	0.02	0.025	0.03		0.035	0.04	0.045	0.05	0.055	0.06
1,500	42	61	80	99	116		133	150	166	181	196	210
1,000	41	60	78	96	112		128	142	157	171	184	196
900	41	60	78	95	111		126	140	154	168	180	192
800	41	59	77	94	109		124	138	151	164	176	187
700	41	59	76	92	107		121	134	147	159	170	181
600	40	58	74	90	104		118	130	142	154	164	174
500	40	57	73	88	101		113	125	136	146	155	164
400	39	55	70	84	96		107	118	127	136	144	152
300	38	53	66	79	89		98	107	115	122	129	135
200	36	49	60	70	78		85	91	97	102	106	110
150	34	45	54	62	69		74	79	83	87	90	93
120	32	42	50	57	62		66	70	73	76	78	81
100	30	39	46	52	56		60	63	65	68	69	71
90	29	38	44	49	53		56	59	61	63	64	66
80	28	36	41	46	49		52	54	56	58	59	60
70	27	34	39	42	45		48	49	51	52	54	55
60	25	31	35	39	41		43	44	46	47	48	48
50	23	28	32	34	36		37	39	40	40	41	42
40	21	25	27	29	31	32		32	33	34	34	34
35	20	23	25	27	28	28		29	30	30	30	31
30	18	21	22	24	24	25		26	26	26	27	27
25	16	18	19	20	21	21		22	22	22	23	23
20	14	15	16	17	17	18		18	18	18	18	19

Sample Size Table

Previously Estimated Proportion of Subscribers Inappropriately Taking Lifeline Service (P)

(N) Number of Lifeline Subscribers	0.01	0.015	0.02	0.025	0.03	0.035	0.04	0.045	0.05	0.055	0.06
17	12	14	14	15	15	15	16	16	16	16	16
15	11	12	13	13	13	14	14	14	14	14	14
13	10	11	11	12	12	12	12	12	12	12	12
11	9	10	10	10	10	10	10	10	10	11	11
10	8	9	9	9	9	9	9	10	10	10	10
9	8	8	8	8	8	9	9	9	9	9	9
8	7	7	7	8	8	8	8	8	8	8	8
7	6	6	7	7	7	7	7	7	7	7	7
6	5	6	6	6	6	6	6	6	6	6	6
5	5	5	5	5	5	5	5	5	5	5	5
4	4	4	4	4	4	4	4	4	4	4	4
3	3	3	3	3	3	3	3	3	3	3	3
2	2	2	2	2	2	2	2	2	2	2	2
1	1	1	1	1	1	1	1	1	1	1	1

1

2 Sample sizes for ETCs with less than 400,000 Lifeline subscribers are calculated pursuant to the following formula: sample size = N/(1+{[N-1]/n}). N is the number of Lifeline subscribers. n is (2.706 * P*(1 – P)) / .000625, where P is the estimated percentage of Lifeline subscribers inappropriately taking Lifeline service. ETCs may choose to calculate their sample sizes using these formulas.

2