";
var eawidth=675;
var eaheight=550;
if (a == 1) { eawidth=725; eaheight=400; intextdi = intextdi + "Our global tracking, ''The Grand Mean Project'' has disclosed tremendous differences between
panels. They are a diverse grouping with considerable purchasing behavior variability. To us it
is not demographics that weigh in as the best measure of panel performance but purchasing
behavior that is most relevant. Our clients are not interested in the problems we experience
with our respondents but rather the data we provide on purchasing. Ask any client if they care
about professional respondents or speeders as much as they do about shifts in purchasing and
you will get the same answer each time. | ";}
if (a == 2) { intextdi = intextdi + " ";}
if (a == 3) { eawidth=745; eaheight=350; intextdi = intextdi + "Managing Within and Between Panel Variability in Online Samples -
In the past five years, increasing scrutiny has been directed towards online data quality issues brought on by panelists engaging in ''satisficing'' behaviors (i.e., doing the minimum to satisfy the survey's requirements) and the ''professional respondent.'' Many researchers began to wonder about the online methodology and whether it truly was a viable means to collect accurate and inexpensive data. Because some researchers could not get findings to replicate from one point in time to another, or when using a different panel supplier as a sample source, many in the industry began to wonder whether they should be using online data collection at all. |
";}
if (a == 4) { eawidth=745; eaheight=350; intextdi = intextdi + "| In online research where we function in a non probabilistic environment, it is at times difficult to identify and define problematic respondents. Much discussion has been centered on respondent behaviors that are considered undesirable; speeders, professionals, and satisficers have all been given considerable space in the market research literature. We propose here a method of blending these behaviors into a single index to assist researchers in rejecting the 'worst' respondents. These respondents provide us with information that does not reflect the balance of the data collected and thus represents a quality challenge. We must arrive at strategies that identify behaviors that bias our data. |
";}
if (a == 5) { intextdi = intextdi + "THE METRICS BEHIND QMETRICS™
We identify six types of ''troubling'' behaviors. Three of these behaviors are considered 'question execution errors' while the other three are referred to as 'aberrant.'
Errors in execution are responses that we believe are wrong. They either reflect intent, confusion by the respondent or inattentiveness. They are captured as a metric by specific questions within the questionnaire. We refer to them as ''failure to follow instructions'' and ''inconsistent responses''.
| | 6 quality metrics are: | | 1. Inconsistency – Happy with standard of living / not happy with standard of living | | 2. Inconsistency – price over brand / brand over price | | 3. Speeders | | 4. Straight-liners | | 5. Failure to Follow Instructions | | 6. Participate in online surveys daily |
";}
if (a == 6) { eawidth=745; eaheight=350; intextdi = intextdi + "| Straight-liners are survey participants who have little variation in their responses. There are some respondents who have no variation. But many more, however, indicate only very little variation. That variation is computed as the standard error around a set of similar questions. In the case of the standard questionnaire, this was based on 31 opinion questions. The standard errors for respondents varied between zero to over three on a seven point scale. Having a standard error below 1 was used to define straight-liners |
";}
if (a == 7) { eawidth=745; eaheight=810; intextdi = intextdi + "| Optimizations | | Panels | Optimum | Average | Expected (1 SE) | Inherent (1 SE) | | M8 | 24% | 33% | | | | M17 | 26% | 33% | | | | M12 | 5% | 34% | | | | Root Mean Square Error | 0.40% | 2.36% | 8.31% | 2.45% | | Panels | Optimum | Average | Expected (1 SE) | Inherent (1 SE) | | M8 | 0% | 33% | | | | M13 | 91% | 33% | | | | M16 | 09% | 34% | | | | Root Mean Square Error | 3.6% | 7.8% | 8.3% | 2.4% | | Panels | Optimum | Average | Expected (1 SE) | Inherent (1 SE) | | M10 | 8% | 33% | | | | M13 | 66% | 33% | | | | M16 | 27% | 34% | | | | Root Mean Square Error | 1.6% | 12.3% | 8.3% | 2.4% |
|
Reliability and consistency of samples can be improved by combinations of panels (Blending
Solutions).
Optimization models improve blending methods, data is needed within each market to create the baselines so that these models can be employed. |
";}
if (a == 8) { eawidth=725; eaheight=350; intextdi = intextdi + "| The Chi Square test indicates the likelihood that two distributions of values are the same. It is a collective test of consistency based on variation. It tends to emphasize specific differences. We use two measures one against the overall average values (internal reference) and other against the Grand Mean. The internal reference measures the relative consistency of the series. The external reference is used to determine if any variation is moving toward or away from an overall measure of effectiveness. |
";}
if (a == 9) { intextdi = intextdi + "Cost Effective Online Data Collection
- AFFORDABLE PROGRAMMING AND HOSTING!
- Online sampling professionals with no panel affiliation. We aggregate our purchases and pass along all the savings to you.
- With each bid, we provide you with three sample prices from independent panel houses. We keep the costs competitive, always in tune with market fluctuations and advise you on the strengths and weaknesses of the providers.
- Online programming and hosting by market researchers, for market researchers.
- Our software, I.S.I.S., is open frame for flexibility, which means whatever you need we can program and deliver.
- Our programming and hosting is backed by a senior project team whose experience is the envy of the industry.
- Our programmers have generated literally thousands of complex studies with a flawless record of excellence.
- Over 15,000 data collection projects completed.
- Let us be your Panel Analysts.
|
";}
if (a == 10) { eawidth=850; eaheight=675; intextdi = intextdi + " ";}
intextdi = intextdi + "
 ";
//alert(intextdi);
adpSize('template01',eawidth,eaheight);
adpLocation('template01',10,10,0);
adpContent('template01', intextdi);
adpShow('template01');
}
function erichide()
{
adpHide('template01');
}
function askme(a)
{
var intextdi;
intextdi = "";
intextdi = intextdi + " ";
var eawidth=375;
var eaheight=325;
var easubject = "";
if (a == 1) { easubject = "The Grand Mean Project™"; }
if (a == 2) { easubject = "Optimum Blending Solutions™"; }
if (a == 3) { easubject = "Consistent Track™"; }
if (a == 4) { easubject = "CPR (Consistency, Predictability, Reliability)™"; }
if (a == 5) { easubject = "QMetrics™"; }
if (a == 6) { easubject = "Crop Duster™"; }
if (a == 7) { easubject = "ourselves"; }
if (a == 8) { easubject = "our quality control measures"; }
if (a == 9) { easubject = "our online data collection"; }
if (a == 10) { easubject = "our phone data collection"; }
if (a == 11) { easubject = "our white papers"; }
if (a == 12) { easubject = "research on research"; }
intextdi = intextdi + "| What would you like to ask us about " + easubject +"? |
For a response, please provide your email address below:
";
intextdi = intextdi + " |
 ";
//alert(intextdi);
adpSize('template01',eawidth,eaheight);
adpLocation('template01',500,175,0);
adpContent('template01', intextdi);
adpShow('template01');
}
function eamail(a)
{
var ear1 = document.getElementById("email");
var ear2 = document.getElementById("message");
//alert(ear1.value);
if (ear2.value.length < 2) { alert("Please enter your question below"); return;}
if (ear1.value.length < 2) { alert("Please enter your email address below"); return;}
if (checkValidation(ear1) == true)
{
document.form1.email.value = ear1.value;
document.form1.message.value = ear2.value;
document.form1.p.value = a;
document.form1.submit();
adpHide('template01');
}
}
function checkValidation ( addressField ) {
if ( stringEmpty ( addressField.value ) )
alert ( "Error! There is no E-Mail address entered" );
else if ( noAtSign ( addressField.value ) )
alert ( "Error! The E-Mail address does not contain an '@' character" );
else if ( nothingBeforeAt ( addressField.value ) )
alert ( "Error! An E-Mail address must contain at least one character before the '@' character" );
else if ( noLeftBracket ( addressField.value ) )
alert ( "Error! The E-Mail address contains a right square bracket ']',\nbut no corresponding left square bracket '['" );
else if ( noRightBracket ( addressField.value ) )
alert ( "Error! The E-Mail address contains a left square bracket '[',\nbut no corresponding right square bracket ']'" );
else if ( noValidPeriod ( addressField.value ) )
alert ( "Error! An E-Mail address must contain a period ('.') character" );
else if ( noValidSuffix ( addressField.value ) )
alert ( "Error! An E-Mail address must contain a two or three character suffix" );
else
return (true);
return ( false );
}
function linkCheckValidation ( formField ) {
if ( checkValidation ( formField ) == true ) {
alert ( 'E-Mail Address Validates OK' );
}
return ( false );
}
function stringEmpty ( address ) {
// CHECK THAT THE STRING IS NOT EMPTY
if ( address.length < 1 ) {
return ( true );
} else {
return ( false );
}
}
function noAtSign ( address ) {
// CHECK THAT THERE IS AN '@' CHARACTER IN THE STRING
if ( address.indexOf ( '@', 0 ) == -1 ) {
return ( true )
} else {
return ( false );
}
}
function nothingBeforeAt ( address ) {
// CHECK THERE IS AT LEAST ONE CHARACTER BEFORE THE '@' CHARACTER
if ( address.indexOf ( '@', 0 ) < 1 ) {
return ( true )
} else {
return ( false );
}
}
function noLeftBracket ( address ) {
// IF EMAIL ADDRESS IN FORM 'user@[255,255,255,0]', THEN CHECK FOR LEFT BRACKET
if ( address.indexOf ( '[', 0 ) == -1 && address.charAt ( address.length - 1 ) == ']' ) {
return ( true )
} else {
return ( false );
}
}
function noRightBracket ( address ) {
// IF EMAIL ADDRESS IN FORM 'user@[255,255,255,0]', THEN CHECK FOR RIGHT BRACKET
if ( address.indexOf ( '[', 0 ) > -1 && address.charAt ( address.length - 1 ) != ']' ) {
return ( true );
} else {
return ( false );
}
}
function noValidPeriod ( address ) {
// IF EMAIL ADDRESS IN FORM 'user@[255,255,255,0]', THEN WE ARE NOT INTERESTED
if ( address.indexOf ( '@', 0 ) > 1 && address.charAt ( address.length - 1 ) == ']' )
return ( false );
// CHECK THAT THERE IS AT LEAST ONE PERIOD IN THE STRING
if ( address.indexOf ( '.', 0 ) == -1 )
return ( true );
return ( false );
}
function noValidSuffix ( address ) {
// IF EMAIL ADDRESS IN FORM 'user@[255,255,255,0]', THEN WE ARE NOT INTERESTED
if ( address.indexOf ( '@', 0 ) > 1 && address.charAt ( address.length - 1 ) == ']' )
return ( false );
// CHECK THAT THERE IS A TWO OR THREE CHARACTER SUFFIX AFTER THE LAST PERIOD
var len = address.length;
var pos = address.lastIndexOf ( '.', len - 1 ) + 1;
if ( ( len - pos ) < 2 || ( len - pos ) > 3 ) {
return ( true );
} else {
return ( false );
}
}
|